the development of student occupational … · second, in recognition of those whose immense...

LEARNING WHAT I WANT TO BE WHEN I GROW UP:

THE CONTEXTUAL EFFECTS OF SCHOOLS

ON STUDENT LOCUS OF CONTROL AND

THE DEVELOPMENT OF STUDENT

OCCUPATIONAL ASPIRATIONS

By

Adrienne L. Corn

Dissertation

Submitted to the Faculty of the

Graduate School of Vanderbilt University

in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

in

Leadership and Policy Studies

May, 2013

Nashville, Tennessee

Approved:

Professor Thomas M. Smith

Professor Bruce Barry

Professor Mark D. Cannon

Professor William R. Doyle

iii

This dissertation is dedicated my husband, Darwin

who shares my passion for creating something from nothing and whose love, support, wisdom, perspective and brilliance

made it possible for me to transform these once blank pages into a doctorate. Thanks for making the journey worth every step—and for walking alongside me.

****

The Tuesday Before Spring by Adrienne L. Corn

Cool light of new snow under clouded morning skies is one of my favorite shades of light. I crawl out from under my own pile of blanketed white to open the wooden slats of the bedroom window blinds and welcome the chilly quiet into the room, melding with the landscape as I slip back into the warmth between brown sheets and green cased pillows under a puffed down comforter. I lay on one side, facing the dawn as it eases into the sky with a neutral palate of muted grays, the sun nestled on the other side of its own layer of blankets, seeming none too eager to leave the cover of clouds, even though it is the end of March and the Tuesday before Easter. This morning, we hibernate for just a little bit longer, letting the last of winter linger. I have become accustomed to the quiet and the stillness of this cold season, lengthening beyond its traditional stay on the southern calendar, my own deadlines long past like the vacation plans and the gardening projects all waiting on the other side of graduation; suspended time familiar to me like the only friend who could understand this Siberian journey, alone with my work, my sole endeavor, less now one of passion than survival in an inhospitable landscape. But I have persisted, like the tedious act of making fire from manure and bits of precious hay, twigs turning ever more quickly between hands rubbed raw from friction in hope of a spark, at first, and then a roaring fire, offering light and warmth to this frozen soul, and once fostered into a full blown blaze, ending my decade-long winter of work, beckoning spring. Still, I linger over the remnants of this time before they, too, are gone, carefully double checking that the words and charts and graphs I once cursed as I labored over their creation are within the acceptable margins, the citations, last name and year, in APA style; futzing over the cover sheet my committee will sign, my name alongside four, more experienced travelers of this terrain, no longer alone; verifying that my graduation hood has been ordered, and the title of my dissertation to be read aloud amidst a backdrop of pomp and circumstance and amassed family and friends is printed clearly and correctly on the paperwork to be submitted, with the faith that when I defend, three days from now, it will be a Good Friday, indeed.

****

Isaiah 46:4

iv

ACKNOWLEDGEMENTS

Few achievements are ever the work of a single individual. In my case, this endeavor is

attributable to three groups of people: those who afforded me the opportunity to prove myself,

those who supported my work throughout the doctoral process and those who contributed to the

successful completion of my degree.

First, the deepest appreciation to those who opened the door to the doctoral degree: Dr.

Gale H. Roid, who saw in me the potential to bridge the world of work and academia and

brought me to Vanderbilt; Dr. Dennis G. Hall, who gave me the benefit of the doubt and allowed

me to prove my potential; Dr. James Guthrie, who offered me residence within his department

and a pathway to my degree; and Dr. Mark Berends, who wisely advised me through the first

few rough years. You all have my enduring gratitude for the time you gave me in each meeting,

your words of counsel and your ongoing encouragement. It is my hope that in this work and my

completion of this degree you find assurances that your faith in me was worthwhile.

Second, in recognition of those whose immense friendship sustained me throughout this

journey: Dr. Dorothy Marcic, Dr. Doug Grisaffe, Dr. Sanjukta Kusari, Dr. Neta Moye, Dr. Dick

Daft, Dr. Elizabeth Lingo, Dr. Norma Anderson, my Vandy Cohort, Dwight and Dr. Linda

Marable, Dr.’s Kristie and Cris Hochwender, Walt and Sally Whittaker, Joy Lohmeyer, Heather

Grisaffe, Shelia Quillen, Jill Dahrens, Andrea Iniguez, Mary Herring, Tim Young, Srini

Balasubramanyam, Janet McGinness, Dr. Estelle Lau, Scott Lumish, the Moys’, Garrick’s,

Aust’s, Weckesser’s, Dr. Li Zhang, Dr. Holly Cumming, June Ahern, Julie Kasiewicz, Dr. Doug

and Summer Williamson, the BarreAmped women, Jeanne Sullivan, Axel and Marita Schultze,

my Facebook friends, and all of those unmentioned but not unimportant. Thank you for

v

commiserating, rejoicing, engaging, praying, laughing, encouraging, problem-solving, fixing,

and otherwise keeping me sane and relatively healthy as I pursued something much more

difficult than I ever imagined it would be. To my parents, Preston and Kathy Corn: my deepest

gratitude for your love and passing along both your faith and your smart genes—because of these

things, the Blue Ribbon is always within reach. To Grandma Twila: thanks for instilling within

our family an abiding appreciation for education and the continued pursuit of knowledge. To the

rest of my extended, fantastic family both in the US and Canada who let me stay in your homes,

borrow your children for hugs and playdates, cry on your shoulders, eat your food and take your

love: I am proud to be related to such a creative, intelligent, fun and diverse group of people.

Finally, a tribute to those who were instrumental in sustaining the momentum toward the

finish line of this marathon so that I could successfully enter the ranks of Vanderbilt Ph.D.’s.:

Rosie Moody, Renee Morgan, Kelly Christie and Liz Leis, who tirelessly assisted me in solving

any and all administrative problems I faced; the brilliant professors who kindled my enthusiasm

for my ideas and those who successfully taught this word girl sophisticated quantitative methods

and instilled a new love for the language of numbers and story of data. The highest tribute goes

to my committee: Dr. Mark D. Cannon, whose insight and sharp wit restored my perspective

when I was lost; Dr. Bruce Barry, whose keen ability to question the unquestionable and bridge

theory and methodology always left me hungry to dig deeper; Dr. William R. Doyle who

encouraged me to continue to move forward undaunted even after multiple setbacks; and Dr.

Thomas M. Smith, who taught me hierarchical linear modeling, agreed to chair my dissertation

committee when I needed a champion, and challenged me to develop and deliver my most

rigorous work to date. I am ever indebted to all of you for your commitment to my doctoral

process and for my success in this achievement of a lifetime.

vi

TABLE OF CONTENTS

Page

DEDICATION ................................................................................................................. iii ACKNOWLEDGMENTS ................................................................................................ iv LIST OF TABLES ............................................................................................................ ix LIST OF FIGURES .......................................................................................................... xi Chapter

I. INTRODUCTION ............................................................................................1 Need for Study ..................................................................................................1 Purpose of Study ...............................................................................................3 Research Interest ...............................................................................................4 Contribution ......................................................................................................5 Overview ...........................................................................................................5

II. REVIEW OF THE LITERATURE ..................................................................6 Educational and Occupational Aspirations and Their Importance ...................6 Linkages Between Occupational and Educational Aspirations ......................11 Role of Schools in the Development of Aspirations .......................................18 The Problem of Misalignment ........................................................................30 Importance of Locus of Control ......................................................................39

Externals ...................................................................................................44 Internals .....................................................................................................45

Schools Impact on Student Locus of Control .................................................46 Importance of Study ........................................................................................51

III. RESEARCH MODEL ....................................................................................52 Theoretical Framework and Conceptual Model .............................................52 Occupational Aspirations ................................................................................53 Locus of Control .............................................................................................54 Academic Tracking, Effort, and Achievement ...............................................55 Diagram of Theoretical Model ........................................................................56 Diagram of Final Empirical Model and Implications .....................................59 Research Question and Hypotheses ................................................................61

Research Question ....................................................................................61 Hypotheses ................................................................................................61

vii

IV. DATA AND METHODS ...............................................................................64 Data Set Sample: NELS:88 .............................................................................64 Methodology ...................................................................................................65 Treatment of Data ...........................................................................................67

Missing Values ..........................................................................................67 Multiple Imputation ..................................................................................68 Weighting ..................................................................................................68

Variables .........................................................................................................69 General Treatment ....................................................................................69 Dependent Variables .................................................................................69 Independent Variables ..............................................................................79

Descriptive Statistics .......................................................................................96 HLM Models ...................................................................................................97

Model 1: Growth Model for Student Locus of Control ............................98 Model 2: Growth Model for Student Occupational Aspirations ...............98 Model 3: Growth Model for Student Academic Achievement .................99 Model 4: Full Growth Model for Student Occupational Aspirations .....100

Mediation Testing .........................................................................................101 Mediation Model Sets ...................................................................................103

School Effects on Occupational Aspirations via Student Locus of Control (Model Set 1) ......................................................................103 School Effects on Student Occupational Aspirations via Academic Achievement (Model Set 2) .................................................................114 School Effects on Student Achievement via Student Locus of Control (Model Set 3) ...........................................................124 Full Mediation Model: School Effects on Occupational Aspirations via Locus of Control and Achievement (Model Set 4) ........................135

Significance and Effect Sizes of Mediation ..................................................138

V. ANALYSIS ...................................................................................................140 Dependent Variables and Growth Over Time ..............................................140 Mediation Models .........................................................................................147

School Effects on Occupational Aspirations via Student Locus of Control (Model Set 1) ..........................................................................148 School Effects on Student Occupational Aspirations via Academic Achievement (Model Set 2) ................................................154 School Effects on Student Achievement via Student Locus of Control (Model Set 3) ......................................................................158 Full Mediation Model: School Effects on Occupational Aspirations via Locus of Control and Achievement (Model Set 4) ........................162

Student Level Characteristics .......................................................................163 Summary of Findings ....................................................................................164

viii

VI. CONTEXTUALIZATION OF RESULTS ...................................................169 The Problem of Misalignment and Social Learning Theory ........................169 Policy Implications and Recommendations ..................................................175 Limitations of the Study and Opportunities for Future Research .................184 Final Words ...................................................................................................186

Appendix

LITERATURE MATRIX ............................................................................188

REFERENCES .............................................................................................193

ix

LIST OF TABLES

Table Page

1. Composite Variable Items for Student Locus of Control .....................................71

2. Variable for Occupational Aspirations in 8th Grade ............................................74

3. Variable Changes for Occupational Aspirations in 10th and 12th Grades .............75

4. Prestige Score Differences Between BY and F1/F2 for Occupational Aspirations ...................................................................................77

5. Variable for Student Occupational Aspirations ....................................................78

6. Variable for Student Academic Achievement ......................................................79

7. Variable for Parental Occupation ..........................................................................84

8. Variable for School Level SES .............................................................................86

9. School Climate Composite Variable Items ...........................................................88

10. Variable for School Level Counseling Department ..............................................92

11. Variable for Analyzing Growth Over Time ..........................................................93

12. Sample of Restructured Data Showing Data for Each Wave in the Analysis ......94

13. Complete Variable Table ......................................................................................95

14. Descriptive Statistics .............................................................................................96

15. Descriptive Statistics for ANOVA’s ...................................................................140

16. Results for Individual ANOVA Models ............................................................145

17. Results for Model 4: Full Growth Model ..........................................................147

18. Results of all Mediation Model Sets ...................................................................149

19. School Contextual Effects for Model Set 1 ........................................................151


x


22. School Contextual Effects on Dependent Variables From Full Model ..............163

23. Effects of Student Level Characteristics .............................................................165

24. Summary of Hypotheses ..................................................................................... 167

xi

LIST OF FIGURES

Figure Page

1. Theoretical Model During Early Stage of Identification Between

Self and Occupation (i.e., Middle School) .........................................................56

2. Dynamic Theoretical Model Illustrating the Effects of Tracking and Academic Achievement on Locus of Control and Occupational Aspirations Over Time .......................................................................................58

3. Dynamic Model of Full Mediation Between School Context Variables and Occupational Aspirations via Locus of Control and Academic Achievement .....................................................................................60

4. Baron & Kenny (1986) Mediation Models .........................................................102

5. Path c ...................................................................................................................102

6. Path a ...................................................................................................................103

7. Paths b and c' .......................................................................................................103

8. Sample of Change in Occupational Aspirations From 8th to 12th grades ............142

9. Sample of Change in Locus of Control From 8th to 12th Grades ........................143

10. Sample of Change in Academic Achievement From 8th to 12th Grades .............144

1

CHAPTER I

INTRODUCTION

Need for Study

The development of student occupational aspirations is particularly relevant to educators

and policy makers interested in enhancing the educational and eventual occupational outcomes

of students. Recent research has revealed that desired educational and subsequent occupational

outcomes for students are increasingly disappointing, with less that 45% of the US population

ages 25-34 having acquired a college degree (J. M. Lee & Rawls, 2010, p. 28), even as the

number of high school students planning to attain a college degree has increased steadily, with

current college aspiration rates at over 80% (Goyette, 2008). The rise in misalignment between

educational and occupational aspirations and attainment (Csikszentmihalyi & Schneider, 2000;

Rosenbaum, 1998, 2001; Schneider & Stevenson, 1999) has highlighted the need for schools to

address this disparity. While educational research has placed a great deal of emphasis on

understanding how schools impact educational aspirations and attainment, given the increasing

misalignment and its subsequent effect on both the individual and society, as well as the

significant relationships between occupational aspirations and both educational and occupational

attainment (Blau & Duncan, 1967; Csikszentmihalyi & Schneider, 2000; Gottfredson, 1981;

Goyette, 2008; Inoue, 1999; Plucker, 1998; Robertshaw & Wolfle, 1980; Rosenbaum, Miller, &

Roy, 1996; Schneider & Stevenson, 1999; Sewell, Haller, & Portes, 1969; Wong, 1997a), a

renewed focus on understanding how schools impact the formation of student occupational

2

aspirations is warranted (Cebi, 2007; Kenny, Blustein, Chaves, Grossman, & Gallagher, 2003;

Kerckhoff, 2000b; Plucker, 1998).

Indeed, the development of occupational aspirations is a critical step for students, as their

aspirations guide their educational and occupational choices toward occupational attainment.

How and when student occupational aspirations are formed has long received cross disciplinary

attention ranging from sociological perspectives on occupational aspirations being significant in

status attainment and the social reproduction cycle (Blau & Duncan, 1967; Bourdieu & Passeron,

1977; Haller & Portes, 1973; Sewell et al., 1969); psychological research analyzing individual

and family level characteristics in relation to aspirations (Hansen & McIntire, 1989; Kerckhoff &

Huff, 1974; McClelland, 1990); and revealing adolescence as the most influential period for

aspirational development and occupational commitment (Holland, 1985; Marcia, 1987; Trice &

McClellan, 1993); social psychology addressing the effects of personal attributes such as self-

efficacy and motivation on aspirational development (Bandura, 1986; Gottfredson, 1981; Lent,

Brown, & Hackett, 1996; Lent, Brown, & Larkin, 1986); and educational research analyzing

aspects of schools in relation to aspirations, including school type (Gamoran, 1996; Rumberger

& Palardy, 2004), locale (Cook et al., 1996; Gamoran, 1996), curriculum (Kerckhoff, 2000a),

student supports (Kenny et al., 2003), and the effects of education policy (Orfield, 1997; Stern,

2009), vocational education (Kerckhoff, 2000a; Rosenbaum, 2001) and school counselors

(Paisley & Hayes, 2003; Lee, Daniels, Puig, Newgent and Nam, 2008).

As noted by Helwig (2004), “career education and its consequent career development

happen everywhere, all the time” (p. 56). As students spend a significant portion of their lives in

schools during developmental stages which are in part associated with occupational exploration,

identification, and aspiration, refining their ideas about “who to be . . . and what occupational

3

direction to pursue” (Marcia, 1987, p.166), it becomes important to examine more closely the

role of school context in the development of student occupational aspirations. Although the

existing literature documents the direct effects of specific school characteristics on the formation

of student occupational aspirations, an examination of how the developmental context of schools

(Kenny et al., 2003) may indirectly affect student’s formational process over time (Helwig, 2004;

Rindfuss, Cooksey, & Sutterlin, 1999) may provide greater insight into how educators can

address the issue of misalignment.

Purpose of Study

Both educators and policy makers seek positive outcomes for students, but require

empirical data upon which to base policy and curriculum decisions. The purpose of this study is

to examine the interaction between the environment of schools and the agentic choices of

students at a time when they are developing and committing to occupational aspirations which

may significantly affect both their educational and occupational outcomes. To do so, the

theoretical framework of this study builds on: (a) prior educational research that acknowledges

this interaction (Csikszentmihalyi & Schneider, 2000; Kerckhoff, 1972; A. M. Ryan & Patrick,

2001; Seifert, 2004); (b) social psychological research that asserts the importance of student

motivation in the development of aspirations, specifically the significant importance of how

much control students feels over their lives (Gottfredson, 1981; Hanson, 1994; Mau & Bikos,

2000; Mortimer, 2000; Mortimer, Vuolo, Staff, Wakefield, & Xie, 2008; Mortimer, Zimmer-

Gembeck, Holmes, & Shanahan, 2002; Wang, Kick, Fraser, & Burns, 1999); and (c) the

educational research which reveals that characteristics of schools impact students, and

specifically, student motivation and sense of personal control (Bradley & Gaa, 1977; Eccles &

4

Midgley, 1989; Eccles & Roeser, 2009; Plucker, 1998; Stipek, 1980), along with the career and

vocational development research which posits that career education is a constant (Helwig, 2004).

Based on this theoretical framework, this study hypothesizes that one way the context of schools

may impact student occupational aspirations is indirectly, by impacting student locus of control.

This study contributes to the educational research in two important ways: first, by analyzing

whether locus of control is a mechanism by which schools have an effect on student occupational

aspirations and second, by examining the effect of schools on locus of control and occupational

aspirations over time.

Research Interest

The research interest in this study is whether there is a contextual effect of schools, and

specific school characteristics, on student occupational aspiration. As such, the research in this

study is focused on the following questions:

Is it possible that characteristics of schools have a significant impact on student

occupational aspiration indirectly via a mediating student characteristic—specifically student

locus of control?

1. Do school characteristics (i.e., academic press, school climate, counselor supports,

SES) directly impact student occupational aspirations?

2. Do the characteristics of schools directly impact student locus of control?

3. Does student locus of control directly impact student occupational aspirations?

4. Given these answers, does student locus of control mediate the relationship between

the characteristics of schools and student occupational aspirations?

5

Contribution

This research provides an important contribution to the literature by examining a

previously unstudied relationship between school environment and student occupational

aspirations via student locus of control and analyzing whether locus of control is a mediating

mechanism by which the context of school impacts the formation of student occupational

aspiration over time.

Overview

• Chapter I outlines current problem, purpose of study, and research questions as well as

the contribution to the existing literature.

• Chapter II provides an overview of the literature regarding the importance of educational

and occupational aspirations, the role of schools in their development, the issue of misalignment,

the role of locus of control in occupational aspirations and how schools may influence student

locus of control and thus student occupational aspirations and eventual outcomes.

• Chapter III provides a discussion of the theoretical framework, hypotheses based on the

literature, and expected results from the analysis.

• Chapter IV discusses the dataset, methodology, measures, and models.

• Chapter V provides an analysis of the results of the study in relation to the hypotheses.

• Chapter VI discusses the results of the study in relation to the theoretical framework, the

limitations of the study, and opportunities for future research as well as policy considerations

stemming from the study.

6

CHAPTER II

REVIEW OF THE LITERATURE

Educational and Occupational Aspirations and Their Importance

Educational and occupational aspirations have been defined by researchers as a person’s

orientation toward particular academic or career goals and the desire to obtain those goals

(Gottfredson, 1981; Hansen & McIntire, 1989; Rojewski, 1996). Educational and occupational

aspirations foster action and persistence when connected to pathways and programs meant for

positive outcomes (Paul, 1997a) and are important at both the individual and societal levels. For

the individual, aspirations form foundational pathways toward future attainment of education,

skills, and jobs. Aspirations serve as a map for future choices, guiding individual agency toward

a desired outcome. For US society, educational attainment and transition into the workforce are

what create economic sustainability (Wong, 1997b). Development of higher aspirations as a

path toward meeting future workforce demands is a necessity in a postindustrial age that requires

highly skilled workers (National Commission on Excellence in Education, 1983; Weiss, 2003).

In addition, aspirations impact individual status attainment, which in turn shift the overall social

stratification.

Due to their importance, research on occupational and educational aspirations has a long

history in sociological and educational literature. One of the earliest studies to examine the

importance of occupational and educational aspirations specifically focused on the process of

formation (Sewell et al., 1969). Status attainment theory explains the formation of aspirations as

the result of social stratification (I. H. Lee & Rojewski, 2009). Sewell et al. (1969) hypothesized

that aspirations are formed out of two sets of influences:

7

1. that of significant others such as parents, friends, and teachers; and

2. that of individual assessment of future potential based on past performance (Haller &

Portes, 1973).

Known as the Wisconsin Model of status attainment, Sewell et al. (1969) estimated the effects of

these influences on the formation of individual educational and occupational aspirations by

employing path analysis with a longitudinal set of data collected from male graduates of

Wisconsin high schools in 1957, with a follow-up in 1964. With a total sample of 929 subjects,

researchers analyzed the impact of socioeconomic status (SES), cognitive ability, academic

performance, and the influence of significant others. Significant others’ influence was measured

by a summed score of four variables which included whether or not the student was encouraged

to go to college by his mother, his father, his teachers and whether or not his friends planned to

attend college. The resulting estimation supported the hypothesis that significant others and SES

had a direct effect on the formation of aspirations. The Wisconsin model was one of the first to

reveal the importance of social psychological influences (i.e., the influence of others on students’

thoughts, feelings or behaviors) on the development of aspirations and was formative in setting

the stage for understanding what contributes to the overall development of aspirations.

Building on past theory and research, Mau and Bikos (2000) also examined how clusters

of variables predicted the formation of educational and occupational aspirations. Using the

National Educational Longitudinal Study data from 1988-1994 (NELS:88) third wave, these

researchers used regression analysis to analyze the impact of personal/psychological

characteristics, school, family, sex, and race on the aspirations of 6,133 students. In keeping

with prior research, Mau and Bikos found that personal characteristics such as race and sex are

highly predictive of the level of aspiration formed by students, although more so for education

8

than for occupation. Likewise, school variables such as academic track and type of school were

also predictive of the level of aspirations formed by students. In other words, minority, male, or

students in a general academic track or in a public school will have predictably lower aspirations

than non-minority, female, or students in a higher academic track or attending a private school.

Personal variables such as self-efficacy, locus of control, and self-concept were also significantly

predictive of the levels of aspirations students formed: those students with higher self-efficacy,

internal locus of control and greater self-concept predicted higher aspirations than students with

low self-efficacy and self-concept scores and an external locus of control.

The formation of aspirations is a critical step for young people in that aspirations help

guide choices toward eventual educational and occupational attainment. Aspirational formation

is especially important in the current social milieu. The need for such aspirations and for young

people to decide what they want to be when they grow up is a relatively modern phenomenon

(Csikszentmihalyi & Schneider 2000), coinciding with the rapid increase in industrialization and

technology (Choy, 2002; Goyette, 2008; Goyette & Mullen, 2006). In the span of roughly two

and a half generations, predictable occupational roles defined by division of labor, familial ties,

or social class have ceased to be the norm (R. B. Lee & DeVore, 1975; Shanahan, 2000). While

aspiring to an occupation outside of those culturally prescribed was once irrelevant, such

occupational aspirations are increasingly requisite to ensure economic self-sufficiency in the

complexity of the 21st century marketplace (Mortimer et al., 2008). Modell (1989) referred to

this process of having to choose between options as “the injection of increasing volition into the

youthful life course” because of a decoupling of “determinate sequences and intervals” (pp. 332-

333) which were previously inevitable. According to Mortimer et al. (2008), “it may seem

difficult, if not impossible, for teenagers to prepare themselves for the unforeseeable future labor

9

market” (p. 2), since they are not sure what they will do, “which role models, if any are valid . . .

what expectations are realistic, what skills are useful or what values are relevant to their futures”

(Csikszentmihalyi & Schneider, 2000, pp. 4-5). Previously predictable roles and livelihoods

have been replaced with a need for young people to form aspirations and to be informed about a

multitude of occupational pathways at a very early age (Trusty, Niles, & Carney, 2005) in order

to gain the educational foundation necessary to pursue the vastly expanded and more technology

driven occupations to which they might aspire. Reinforcing the need for students to be prepared

and the disappearance of predictable occupational roles, the U.S. Bureau of Labor Statistics

(2012) reported that the average baby boomer changed jobs over 11 times between the ages of 18

and 46. While there is a necessity for students to form aspirations for overall success in the

modern age, they must also be aware of the need to continually reinvest in the aspiration and

attainment process for a lifetime in which job change is a constant reality (Savickas et al., 2009).

At the societal level, educational and occupational aspirations and attainment are

important because they are tied to the underpinning of the American social structure (Blau &

Duncan, 1967; Haller & Portes, 1973; Haller & Woefel, 1972). According to Blau and Duncan

(1967),

technological progress has created a need for advanced knowledge and skills on the part of a large proportion of the labor force, not merely a small professional elite. Under these conditions society cannot any longer afford the waste of human resources a rigid class structure entails. (p. 431)

Greater mobility across the occupational attainment structures is imperative in that it allows for

increased numbers of people to gain and contribute knowledge, skills and abilities necessary in a

technological age. The importance of mobility is supported by aspirations which are linked to

desires for increased status attainment and social mobility (Blau & Duncan, 1967; Gamoran,

2000; Haller & Portes, 1973; Inoue, 1999; Sewell, Haller, & Ohlendorf, 1970; Sewell et al.,

10

1969; Sorokin, 1927). Research reveals that aspirational development crosses boundaries of

class and race. Students from lower SES and minority populations have high aspirations for their

futures even if they don’t have the same access to resources to achieve these goals as their non-

minority, higher SES counterparts (Merton, 1957; Wang et al., 1999). Similarly, students with

disabilities have aspirations for their futures, albeit somewhat lower than their peers without

disabilities (Rojewski, 1996). What many disadvantaged students recognize is that higher status

occupations net greater social mobility (Blau & Duncan, 1967, p. 56), and this recognition in part

fuels their aspirations.

Although educational and occupational aspirations are vital to the individual and society,

the fundamental development of aspirations remains largely that of the individual student. Given

both the agentic nature of educational and occupational choices and the societal need for students

to foster aspirations in preparation for the future, understanding not only what motivates and

shapes student aspirations, but their impact on future attainment is crucial. Acknowledged as

essential pathways toward the future, whether aspirations actually predict future educational and

occupational attainment has been widely discussed. Early estimations from the Wisconsin model

(Sewell et al., 1969) reveal that aspirations have substantial influence on future attainment. In

later research, Rojewski (1996) argued that aspirations may not be determinants of future

attainment but are important since they represent orientation toward particular attainment, and

that educational aspirations tend to have direct bearing on eventual occupational attainment and

play an active role in determining whether students should pursue or ignore educational

opportunities available to them, especially in high school (Inoue, 1999; Rojewski, 1996).

Echoing early research in which Strong (1953) predicted a .69 correlation between a student’s

first choice expressed occupational interest and their attained occupation 19 years later, more

11

recent analysis of aspirations in the areas of education and occupation has shown that aspirations

are among the most useful predictors of eventual educational and occupations choices (Blau &

Duncan, 1967; Gottfredson, 1981; Marjoribanks, 1985). Research by Schoon and Parsons

(2002) corroborated these early findings, establishing that expressed occupational aspirations are

equal to or better than interest inventories in predicting future occupational attainment. Research

by Paul (1997a) found that career aspirations were important when turning intent into endeavor,

whether that intent is to go on to college or into the working world after high school. Given the

extensive research demonstrating the effect of aspirations on future attainment and society’s need

for students to aspire to and attain higher educational and occupational goals in relation to the

technological era, the importance of aspirations is clear.

Linkages Between Occupational and Educational Aspirations

The educational and occupational structures in the United States are “so closely

intertwined that it is almost impossible to discuss one without discussing the other” (Inoue, 1999,

p. 1; Woefel, 1972). Sociologists have long recognized that in the US, more education often

equates to higher status (Gamoran, 2000; Sorokin, 1927), and research confirms a high

correlation between one’s level of education and income (Wong, 1997a). Indeed, the strong and

significant effects of academic achievement on occupation and earnings increase steadily up to

age 35 (Jencks et al., 1979; Murnane, Willett, & Levy, 1995; O'Rand, 2000). An economic

theory of human capital (Becker, 1964) summarizes these outcomes based on two main tenets:

1. that workers with better educational backgrounds and training fare better in the labor

market and

12

2. nations with larger pools of human capital resources will be more competitive in the

global economy, resulting in higher paying jobs.

Therefore, investing in education yields increased opportunity and earnings for individuals and

helps ensure national economic stability. As noted earlier, educational and occupational

aspirations lay the groundwork for future educational and occupational attainment and their

subsequent individual and collective outcomes.

More specifically, academic achievement links educational and occupational aspirations

and outcomes (Gamoran, 1996; Helwig, 2004; Inoue, 1999; NCES, 2006; Rindfuss et al., 1999;

Rosenbaum, 2001). Research shows that strong academic performance enhances future

occupational opportunities and provides youth with strong advantages in occupational attainment

and earnings (Parcel & Dufur, 2001) while weak academic performance and academic

attainment predicts lower earnings (S. R. Miller & Rosenbaum, 1997). In a longitudinal analysis

examining the determinants of earnings immediately after high school, Rosenbaum (2001) found

that test scores have a strong and significant influence on students’ later earnings rather than

immediate earnings, but the effects of achievement remain highly significant over time. Indeed,

people’s jobs and wages are a primarily a function of their educational credentials and cognitive

skills developed through schooling (M. K. Johnson & Mortimer, 2002). In their research on how

teens prepare for the world of work, Csikszentmihalyi and Schneider (2000) analyzed data from

the second year of the longitudinal Sloan Study of Youth and Social Development on 140 high

school graduates one year after graduation. Utilizing both logistic regression models and path

analysis to understand the effects of academic and background variables on outcomes by

examining the number, type, and level of difficulty of courses taken in the previous year as well

as variables such as parental education, gender, race and social class, results reveal that academic

13

performance is an important predictor of college enrollment, which in turn effects future

occupational opportunities. Researchers also discovered that a key predictor in determining

whether a student will go to college or directly into the workforce is high school GPA. The

types of classes, and specifically the number of math and sciences classes taken in high school,

are significant predictors of college enrollment and whether a student will pursue a four-year

college or a community college. Indeed, research in various disciplines has demonstrated

significant and positive relationships between academic achievement and higher educational and

occupational aspirations (Mau, 1995). Doing well academically by taking challenging courses

and getting good grades offers all students, especially those from challenged or impoverished

environments, greater opportunity to pursue both educational and occupational aspirations.

Occupational aspirations have direct effects on educational aspirations, and are often

fueled by the desire for higher status attainment (Csikszentmihalyi & Schneider, 2000; Inoue,

1999). In the US, occupation is a strong determinant of wealth, lifestyle, and individual status

within their community (M. K. Johnson & Mortimer, 2002). According to researchers, the

public thinks about education primarily in terms of preparation for work, better jobs, and better

opportunities in life (Orfield, 1997). Adolescents, regardless of race, have an overall belief that

if they receive a good education, they will be able to get a good job (Steinberg, Dornbusch, &

Brown, 1992), and that “school was viewed as the vehicle leading to better paying jobs”

(Okagaki, 2001, p. 12). Csikszentmihalyi and Schneider (2000) noted that wealth and material

concerns, specifically “the means to indulge in expensive leisure and consumption habits” (p. 9),

motivates many adolescents toward higher occupational goals. As noted previously, jobs are

increasingly knowledge and technology driven, which requires additional education. Thus,

14

students often aspire to educational goals as stepping stones to their occupational aspirations and,

in turn, their aspirations for enhanced status (Goyette, 2008).

Which occupations qualify as having higher status has been the focus of sociologists such

as Nakao and Treas (1992, 1994) who developed a widely used occupational prestige scale.

Occupational prestige is the general level of social standing enjoyed by those within a specific

occupation (Hauser & Warren, 1997) and prestige scales are often utilized by sociologists and

other social scientists in an effort to understand motivations, conceptions, and behavior related to

work and occupations. Previous large-scale survey research in this area was based on both

occupational titles from census data and a large representative sample regarding public opinion

on the prestige of these occupational titles was carried out by the National Opinion Research

Center dating back to 1947 (Reiss, 1961) and subsequently expounded upon and updated by

researchers throughout the following decades (Duncan, 1961; Hauser & Feathermen, 1977;

Hauser & Warren, 1997). The most current index was developed by Nakao and Treas (1994),

and is widely accepted because it reflects not only updated titles but measured the shifts in public

opinion regarding the prestige of those titles rather than apply outdated public opinion to the

updated occupational categories. Nakao and Treas found that in comparing previous

occupational scales from the 1940s through the 1970s (Duncan, 1961; Hauser & Feathermen,

1977; Reiss, 1961; Stevens & Cho, 1985; Stevens & Hoisington, 1987) to the Nakao-Treas scale

which was developed using the occupational categories found in the 1980 US census and

subsequently updated using the 1990 census occupational categories, that a significant factor in

the difference in prestige scores was recent shifts in public opinion, with 44% of occupational

titles experiencing a statistically significant change (p < 0.05) in prestige score between 1964 and

1989. In general, prestige of occupational titles rose. However, some occupations which had

15

once been comfortably situated at the top of the prestige scale in the previous decades such as

being a banker or a draftsman were now considered less prestigious, while different occupations

such as those in entertainment or technology were becoming more prestigious (Nakao & Treas,

1994).

Yet, higher status occupational aspirations may go unfulfilled due to constrained

opportunities (M. K. Johnson & Mortimer, 2002). I. H. Lee and Rojewski (2009) posited that

“aspirations are formed at an early age by the opportunities or barriers presented to individuals

through external factors such as bias, discrimination, cultural expectations, societal attitudes and

stereotypes based on gender, race and social class” (p. 83). This position points to perhaps a less

obvious but equally important linkage between educational and occupational aspirations.

Inequalities in student background such as SES (Bourdieu, 1973; Kerckhoff, 1976; McClelland,

1990; Sewell et al., 1969), personal ability and even organizational features within and between

schools (V. E. Lee, Bryk, & Smith, 1993) such as quality of instruction (Oakes, Gamoran, &

Page, 1992), interest and commitment of teachers (V. E. Lee & Bryk, 1989), ability grouping

(Gamoran, 2000; Gamoran & Berends, 1987) or access to vocational training (Kerckhoff, 2000a;

Schneider & Stevenson, 1999) can contribute to differentials in student educational outcomes

and have implications for subsequent occupational prospects and career outcomes (M. K.

Johnson & Mortimer, 2002). If inequalities in background are reinscribed in educational

structures, high occupational aspirations may seem unrealistic and consequently students within

constrained opportunity environments may not pursue such aspirations (Hanson, 1994).

Rojewski (2005) noted that students with lower occupational expectations are more likely to

make educational compromises, restricting the range of future educational and occupational

16

options. Lower educational aspirations are linked to low occupational attainment (Inoue, 1999),

correlating to lower levels of socioeconomic status (Blau & Duncan, 1967; Rojewski, 2005).

Another reason why students may not form high educational and occupational aspirations

is related to individual assessment of performance and subsequent opportunity (Sewell et al.,

1969). Social psychological research reveals that earlier achievements enhance self-selection for

future opportunities (Alwin, 1994; Mortimer, 2000), thus students who assess their personal

achievement and find it lacking my also limit their aspirational goals based on what opportunities

they assess will be available to them based on their level of achievement. Often such self-

evaluations are reinforced over time as prior achievement regulates the access to future

opportunities for achievement (O'Rand, 2000). Students modify their aspirations based on

learning experiences and performance outcomes (Lent et al., 1996). In relation to this feedback

loop, aspirations of children vary over time (Kerckhoff, 1974), but become more invariant with

age (Rindfuss et al., 1999). By the end of 12th grade, most student aspirations are crystallized

and more realistic of what they expect to achieve (Helwig, 2004; I. H. Lee & Rojewski, 2009),

even if lower than the point of their original conception (Dennehy & Mortimer, 1993; Rindfuss

et al., 1999).

Research on student perceptions and decision making is exemplified in a two-part study

by Kenny et al. (2003), who led a research team in examining the relationship of students’

perceived barriers and support with subsequent levels of school engagement and vocational

attitudes. The premise of the two-part study was based on previous research establishing that

urban youth face many obstacles in pursuit of education and career goals and due to these

obstacles are more likely to disengage from school and post-school planning. These barriers are

attributed to the lack of resources available to prepare urban youth for success in work or

17

college, thus leaving them with limited skills and limited opportunities. Additional research

focused on “contextual career supports” has been recommended for the purpose of exploring

ways to help individuals overcome perceived barriers. Kenny et al.’s research was designed for

that purpose, and focused on documenting the relationship between perceived educational and

occupational barriers and support to school engagement and career attitudes in urban youth.

Hypotheses for the study included the expectation that perceived barriers would be negatively

associated with school engagement and occupational aspirations, and that kinship support would

be positively associated with these two aspects. Study 1 included 174 students, and Study 2

included 181 students, both samples from public high schools in a large northeastern city. Data

were gathered through questionnaires given to ninth-graders involved in classes specifically

designed to link school to work.

Results from the first study confirmed the hypotheses and reported that where there were

lower levels of perceived barriers, there were higher levels of school engagement and higher

aspirations for future careers. Study 2 set out to understand the relationship of these variables in

greater depth and found that perceived barriers existed outside of the context of school. The

barriers outside of school context, such as poverty and discrimination, had an influence on

engagement within school. Likewise, students who had high career aspirations but believed

those aspirations were not attainable because of perceived barriers, might choose to disengage.

Schools that magnify, or add to, perceived barriers through actions such as default tracking (i.e.,

basing academic track placements on previous tracking rather than consideration of other

variables such as academic performance) increase the risk of student disengagement, which

inhibits or limits the process of exploration and subsequent aspirational formation. Kenny et al.

(2003) noted that perception of support and barriers influence development of adaptive attitudes

18

about the self in relation to school and work, that students in the sample are “shaping their lives

to some extent from their interpretation of environmental events” (p. 147).

While Kerckhoff (2000a) acknowledged that “we have yet to do an adequate job of

accounting for the kinds of linkages between educational and labor force institutions” (p. 51),

Kenny et al.’s (2003) work highlights the inherent linkages between school environments and

mindset and subsequent engagement and outcomes of adolescents. Many years prior to actually

entering the working world, students are interpreting their potential place in that world from their

educational interaction. With factors such as academic achievement, educational environment,

and self-assessment contributing to the development of aspirations, adolescents may not always

aspire to higher levels of educational or occupational attainment even while they desire great

social status or personal wealth because they perceive barriers to attainment. While lowered

aspirations due to perceived barriers are not a desired outcome, even such perceptions indicate

that the linkages between educational and occupational aspirations persist.

Role of Schools in the Development of Aspirations

Based on the importance of, and clear linkages between, educational and occupational

aspirations, understanding the role of schools in the development of those aspirations is critical.

To that extent one must consider the public expectations of schools, their function and

functioning. What are the public expectations of school in relation to preparing students for the

working world? How do educational policy decisions reflect the linkages between educational

and occupational aspirations? Does how schools work impact aspirational formation? How does

the contextual environment of schools—specifically interaction with teachers, curriculum

19

offerings and tracks, availability and access to counselors—affect student aspirations? This

section will discuss each of these questions with a review of relevant literature.

Expectations of schools in relation to preparing students for the world of work varies,

although as Hallinan (2000) pointed out, it is widely accepted that one of schools three main

functions is economic: to train future workers for participation in the labor market and thus

contribute to the country’s economic growth and technological development (p. 153). According

to Savickas et al. (2009) school must do more than prepare students for a future with high

technological requirements; it must also prepare them for one in which they will experience

many changes in their career path. Still, unlike other industrialized countries, the US has a very

loose coupling between education and labor (Kerckhoff, 2000a). American students often leave

school without specific occupational preparation, job skills, or work credentials and are

unprepared for how to enter the labor market after high school (Csikszentmihalyi & Schneider,

2000; Rosenbaum, 2001; Schneider & Stevenson, 1999). In the US system, there are few widely

developed and accepted connections between schools and employers (Paul, 1997a), even where

vocational education exists (Rosenbaum, 2001). In his work on educational and occupational

linkages, Kerckhoff (2000a) noted that without some vocational information and direction,

students will experience more turbulence early on in their careers, resulting in a hit-or-miss

approach to their occupational attainment process.

Although the US education and occupational structures are loosely coupled, education

policy makers acknowledge the linkages between education and occupation, at times with

education policy that overtly addresses the issue. This was the case with educational legislation

enacted in 1994. The goal of the School to Work Opportunities Act of 1994 (STWOA) was to

help students at all levels of K-12 education “obtain the experience, knowledge, and skills

20

required to explore the world of work, develop employment skills, make decisions and to

identify, pursue and attain vocational goals” (McWhirter, Rasheed, & Crothers, 2000, p. 330).

This act was an important piece of legislation, as it placed a high economic value on the

connection between school and work. It also placed a spotlight on this now federally funded

transition process, spurring many researchers to focus on its efficacy.

The Indiana Youth Opportunity Study (IYOS), a longitudinal study that focused on

students’ transition from school to work, surveyed 5,200 students in the 8th, 10th, and 12th grades

across 52 schools in the state of Indiana, along with their parents and counselors, and then

followed the 12th graders for two years after graduation. Begun in 1991-1992 with the initial

survey, the study administered a follow-up survey to the 12th graders two years after graduation

in 1994-1995. The data from the study were used to compare student’s career expectations while

in high school with employment outcomes after high school (Wong, 1997a, 1997b). The

strength of this study lies in its sample size (N = 5200) and its design in including the three

specific groups that most influence adolescent decisions in relation to occupational decision

making: parents, schools (specifically, counselors) and the students themselves. With significant

data on all three groups, the study was able to analyze the role of school in relation to the

adolescent and the reality of the academic decision-making process in relation to the working

world. Perhaps an even greater strength of this study was its longitudinal design to track student

academic choices with outcomes, providing valuable data on the effects of academic program

preparation (or lack thereof) and tracking. This study also briefly touched upon an oft forgotten

population of adolescents—dropouts—and discussed potential issues this population faces based

on its decision to leave school. Critics of the IYOS study have noted that results needed to be

more specific and take into account multiple variables involved in the decision-making process

21

for the students studied, controlling for student attributes such as poor attendance or behavior

which might influence the research results. Critics suggested that if true change is to come from

such research, schools need more than average outcome information; they need specific student

detail on choice and outcome in order to address focused challenges facing the student

populations in their specific schools (Rosenbaum, Miller, & Krei, 1997, p. 90).

The IYOS study also found that in many cases, parents relied on schools to help their

children prepare for the years beyond high school because they themselves lacked the expertise

to do so. Orfield (1997) noted that the educational and occupational level of parents correlated

with a willingness to concede educational decisions to school staff, and counselors were counted

on to help guide their children to college or successful careers. Additionally, students in the

study “reported a serious need for assistance in obtaining information about jobs and seeking

employment. For those whose families have poor information, as many do, the school plays a

vital role” (Orfield, 1997, p. 21). The IYOS findings are echoed by other research, which cites

the importance of the school’s role in providing information on career options and opportunities

(Kelly & Lee, 2001; McWhirter et al., 2000; Mortimer et al., 2002). This is especially true in

minority groups and lower socioeconomic communities where information about options among

parental groups is lacking (Kenny et al., 2003; Okagaki, 2001).

Results from the IYOS reveal that 70% of students went on to pursue their desired path,

whether college or work. For those heading to college, the primary factors in converting intent

to endeavor were high school courses taken (college prep track) and positive perception of

academic efficacy. For those going into the labor force, the primary factors related to converting

intent into endeavor were race (non white) and parents level of education (high school or less).

Even for those who were enrolled in a vocational track, their courses in this track did not assist

22

them in turning intent into action if their goal was anything other than the work world (military

or post-secondary vocational education). For those students who went on to college but did not

persist past the first year (20%), the factors most influencing their lack of persistence were level

of parental education (high school or less), perception of academic efficacy (low grades), and not

taking the appropriate high school courses. Eighty-seven percent of those students ended up

entering the workforce (Paul, 1997b).

Given the IYOS study and other research on actual school to work practices in play after

the 1994 STWOA legislation (Helwig, 2004; Orfield, 1997; Paul, 1997a, 1997b; Rosenbaum,

2001; Wong, 1997b), the question of whether the school to work legislation was an efficacious

policy in terms of developing for students clear and successful transitional pathways from school

to work has since been answered with a negative. Ultimately, the school to work legislation

initiatives to help link school based education with internships and work preparation were slow

to be adopted by states, as the need for developing the institutional linkages between schools and

the labor market was not clearly understood by educators and counselors--the primary source of

information for students choosing courses (Wong, 1997a). Thus, students were left to choose

courses and academic tracks based on perceptions that those courses would lead successfully to

their desired outcomes, despite the fact that for vocational track students, coursework did not

assist them in turning intent into endeavor, nor did college prep track assist those students who

did not persist in college and ended up entering the workforce prematurely. The School to Work

Opportunities Act of 1994 is one example of how policy may be designed to impact the role of

schools and subsequently the goals and choices of students—although the ultimate impact and

outcomes may not reflect its original design.

23

Coursework, and specifically tracking, is another area in which schools play a role in the

development of students’ aspirations. Tracking starts early and is significantly persistent

(Alexander & Entwisle, 2000). Students are placed into ability groups as early as first grade,

launching them into “achievement trajectories that persist” (M. K. Johnson & Mortimer, 2002)

because of the effects that placement in an ability group or track at one grade level has on the

placement at subsequent grade levels (Alexander & Entwisle, 2000; Entwisle & Alexander,

1993). Early tracking is often decided by educators alone, often based on ability or assessed

readiness.

The IYOS study (Orfield, 1997) showed significant statistical correlations between the

educational and socioeconomic level of parents and those students who chose vocational training

programs. Children of parents with less education or lower status jobs were more likely to

choose vocational programs. Race, sex, and academic ability were not significant factors in who

chose these tracks. IYOS indicates that there was also a consistent pattern of less information

and less feeling of confidence and empowerment about program decisions among students and

parents in less educated and affluent families (Orfield, 1997). Clearly, this study indicates that

vocational tracking can be a significant barrier to future opportunity. Often students and parents

have little idea as to how much of a barrier because there is little connection made between the

academic program and future aspirations and even less information available to these students

and parents on the how various academic programs might affect future opportunities.

The IYOS also noted that the effect of tracking on students is long term. In the IYOS

student surveys, most students communicated positive educational and career aspirations, but did

not have the appropriate information to link those aspirations with the necessary steps for

achievement, ultimately affecting post-secondary outcomes. The study found the misalignment

24

between aspirations and preparation crossed class, race, and gender lines (Paul, 1997a, p. 94).

Longitudinal research confirms that academic choices made as early as middle school have a

strong bearing on educational and career development (Rosenbaum, 1998; Trusty et al., 2005)

and that track placement is a mediator between student background and academic achievement

(Kerckhoff, 1995; V. E. Lee & Bryk, 1988). Berends (1992) noted that students in honors

classes were more likely to maintain high expectations than students in other classes, even

among students with similar test scores and initial expectations. Indeed, a significant body of

research confirms that college preparatory tracking is related to higher educational expectations

for attainment (Gamoran & Berends, 1987). Still, research by Hallinan (1996) argued that

tracking is not as stringent as once thought, but that movement between tracks still reflects and

perpetuates student inequalities in race and gender. Hallinan posited that although tracking

changes may be based on academic achievement, after controlling for achievement, the variables

gender, SES, and age impact student risk for “tracking down.” Since academic achievement is

predictive of future occupational opportunity and earnings (Csikszentmihalyi & Schneider, 2000;

S. R. Miller & Rosenbaum, 1997; Parcel & Dufur, 2001; Rosenbaum, 2001) as well as a point

for individual evaluation which may limit aspirational development (Alwin, 1994; Lent et al.,

1996; O'Rand, 2000), tracking is one way schools play a role in educational and occupations

development of students.

As noted in the previous section, student perception of contextual supports or barriers can

affect their levels of engagement (Daniels & Araposthasis, 2005; Klem & Connell, 2004), which

in turn affects level of achievement (Montalvo, Mansfield, & Miller, 2007; Wehlage, Rutter,

Smith, Lesko, & Fernandez, 1989) and sets the groundwork for student aspirations and future

attainment. The 2007-2008 High School Survey of Student Engagement asked a national sample

25

of 6,424 students why they were in school. Fifty-eight percent of those asked responded that

they are required by law to attend, with fewer than 40% responding that they are in school

because they are interested in learning (Yazzie-Mintz, 2009). While overall levels of

engagement are less than ideal given desired student outcomes, such levels are significantly

affected by feeling connected to adults in the school community and whether students felt

teachers and counselors believed in them (Daniels & Araposthasis, 2005). Research by Plucker

(1998) specifically ties supportive school environments to increased student aspirations.

Research by Patrick, Kaplan and Ryan (2007) on the classroom environment and student

motivation and engagement found strong evidence that emotional support such as

encouragement from teachers improved student engagement, focus on skill mastery and feelings

of efficacy (Eccles & Wigfield, 1985; A. M. Ryan & Patrick, 2001). While teacher supports are

defined slightly differently by researchers (Goodenow, 1993; Skinner & Belmont, 1993), it

typically encompasses features such as caring, understanding, and dependability and refers to the

extent students believe teachers value and foster personal relationships with them. Such beliefs

contribute dimensionality to the classroom and student perceptions of the support within that

context (A. M. Ryan & Patrick, 2001; Schoon, 2001). The specific impact of teachers on student

occupational interest is important and increases over time. Helwig (2004) noted that by the time

students reach 12th grade, teachers were reported as having the most impact on occupations

interests, even with students reporting mixed or negative views about the school’s overall career

help or support. Franklin (1995) revealed that student interaction with teachers had a strong

relationship with academic pursuits and perceptions of cognitive development, which are then

used in student self-determination of educational and occupational aspirations. Flowers, Milner,

and Moore (2003) found that African American students who felt that their teacher had high

26

expectations of them had higher educational aspirations but that teachers and counselors need to

communicate high aspirations from the beginning of students’ educational careers.

In addition, teacher quality and pedagogical methods also impacts engagement and

achievement. The academic tasks assigned by teachers to students and the way in which teachers

organize classroom participation structures for students contribute to the social environment and

thus student learning motivations (A. M. Ryan & Patrick, 2001). Work by educational

researchers suggests that instructional content and methods vary systematically, typically

favoring those in the higher level classes and working against those in lower level classes, which

are typically at the bottom of the status hierarchy (Barr & Dreeben, 1983; Gamoran & Berends,

1987; Gamoran & Mare, 1989; Hoffer, 1992; Kerckhoff, 1986, 1993; Murphy & Hallinger,

1989). Thus, the classroom organization within a school as determined by teachers can impact

motivations and aspirational development.

The role of guidance counselors in student trajectories has less prominence in educational

research literature, but has received a great deal of research attention from vocational and career

development and the field of counseling itself. A 2004 review of career development literature

(Hughes & Karp) examined the history and effectiveness of career guidance and counseling,

specifically addressing the work performed by school counselors. Noting that counseling gained

prominence in public schools at the turn of the century as a way of assisting students in aligning

an understanding of self with knowledge, skills and the opportunities available to utilize them,

the authors recognize that this was also the beginning of a trend toward “differentiation” tracking

in public education based on a counselors determination of future opportunities and outcomes for

particular students. Toward the 1970s, a less individualized and more broad based, structured

program known as the Comprehensive Guidance Program model was developed, refined and by

27

the 1980s widely adopted for use in local school districts nationwide. In the last two decades,

guidance has received little emphasis in educational reforms, even with the School to Work

Opportunities Act of 1994. No Child Left Behind legislation has encouraged academic guidance

over vocational or career counseling to coincide with more standards-based outcomes (Carey &

Dimmitt, 2008), with only 8% of schools focused on any form of direct career and vocational

guidance with students (Hughes & Karp, 2004). According to Hughes and Karp, research in the

area of career development confirms that the most often cited activity among counselors is

helping students with course scheduling and college admissions, even though career

development skills training is predictive of adolescents’ goal orientation, directedness, and

confidence in their abilities to pursue chosen career paths (Janosz, LeBlanc, Boulerice, &

Tremblay, 1997; Turner, 2007). Students seeking vocational guidance are often left to fend for

themselves, utilizing what self-help career planning tools may be available within the counseling

center.

Counselors were not considered helpful to the large number of students who required

their assistance in helping them find relevant academic paths. According to the IYOS (Paul,

1997a; Wong, 1997a), counselors had large counseling loads and little knowledge about work

placement or vocational education. The typical course of action was to talk with students about

college, even though there was high recognition for the need to help students that might not go to

college plan and prepare for work roles after high school. This need was not seen to be as

important as college preparation and less than 7% of their time was spent counseling students

about alternatives to college. Yet, even the importance of a college preparatory program was not

wholly understood by students having to choose which program to embark upon. Often

28

decisions about academic tracks were made by default by students with very little information

about the importance and long term effects of their decision.

The IYOS study stated that

educational and opportunity capital available in high school and required for post high school endeavor and lifelong social and economic opportunity, is mediated by academic program. But the decision process, and the way the decision is treated over the high school years, gives little indication of its immediate and long-term importance. (Paul, 1997b, p. 44)

In terms of academic program choice, the study surveyed eighth-grade school counselors who

were asked how they made recommendations for which, if any, high school prep classes (and

thus high school program) a student should pursue. From performance criteria such as grades,

achievement test scores, IQ tests, or parental and student requests, overwhelmingly counselors

made their recommendations based on teacher recommendations, regardless of other factors.

Data show that students relied on the influences of guidance counselors in choosing classes in

preparation for high school and based on that influence, students rarely revisited their curriculum

decision throughout their high school years. This was especially true if there were few

opportunities to visit with high school counselors. The study stated that for “90% of the students

in our sample, there had been no discussion of broad educational and career goals, or of the more

specific issue of program selection for high school, since as far back as the seventh grade” (Paul,

1997b, p. 45).

Research on counseling outcomes reveals significant and positive relationships between

student achievement and occupational development and fully implemented guidance programs

(Borders & Drury, 1992; Fouad, 1995; Gerler, 1985; Lapan, 2004; Lapan, Aoyagi, & Kayson,

2007; Lapan, Gysbers, Hughey, & Arni, 1993; Lapan, Gysbers, & Sun, 1997; Whiston & Sexton,

1998). This research has early beginnings with Caravello (1958), whose longitudinal research

29

revealed that one year after graduation, students who had received counseling assistance were

more likely to have developed aspirational goals and continue toward those goals after high

school. Rothney (1958) found that school counseling had longitudinal effects five years after

leaving high school, with students having higher levels of achievement, expressing and attaining

aspirations, and persisting in their attained aspirations longer than students who did not receive

counseling assistance. Almost 50 years later, McWhirter et al. (2000) echoed these findings in

their more recent research using a intervention model which showed significant increases in

career decision making as well as increases in their expectations for a satisfying career, but their

research showed gains in these areas were not sustainable past the end of the course intervention,

casting doubt on the sustainability of the significant effects of counseling outcomes over time.

While interaction with school counselors may have positive (if not sustainable) student effects,

studies show that a low percentage of students actually seek out counselors to discuss either

academic or vocational planning (Mau, 1995) and many students have little or no access to

counselors at all (CEEB, 1986). Even where there are counselors available, high student-to-

counselor ratios or alternate counselor responsibilities may limit availability of counseling

services to individual students (Boyer, 1983; V. E. Lee & Ekstrom, 1987). Such gaps in

counselor availability has promoted the use of computer assisted career guidance tools that often

require little or no oversight by a guidance counselor, but research on the breadth of their effects

and the sustainability of those effects over time is lacking. School counseling is yet another area

where schools impact student trajectories, either through their interaction with students, or their

lack of it.

In view of research on tracking, student perceptions of contextual supports, actual

contextual effects such as teacher practices and actions of and access to counselors, it becomes

30

clear that schools have an impact on students’ self-assessments, motivations, achievement, and

development of aspirations. Indeed, as much research in these areas has confirmed, schools are

socializing agencies (Alexander & McDill, 1976; Heyns, 1974) that legitimize hierarchies in

which people are allocated (i.e., tracks) (Gamoran & Berends, 1987; Gamoran & Mare, 1989;

Kerckhoff, 1976; Meyer, 1977). Kerckhoff (1993) detailed the cumulative effects of school

organization on both academic achievement and occupational attainment and shows the

significant and long-term consequences of those effects—demonstrating through research that

formal social structures such as those in schools systematically effect student characteristics and

alter outcomes. As noted by Schoon (2001), “the school environment can be either a powerful

shaper or deterrent to the development of children’s aspirations” (p. 125).

The Problem of Misalignment

The issue of how schools impact student aspirations is critically pertinent given the

relatively sudden onset of a current concern with such aspirations: the issue of misalignment

between student aspiration and attainment. According Schneider and Stevenson’s seminal

research (1999), teenagers in the1990s were the most occupationally and educationally ambitious

generation in US history, with 70% of students expecting to receive a college degree as

compared with 30% in 1955. Ongoing research reveals that recent generations of teens are

equally ambitious, if not more so (Reynolds, Stewart, Macdonald, & Sischo, 2006; Yazzie-

Mintz, 2009), with educational expectations steadily rising and 84.5% of students in 2002

expecting to receive a college degree (Goyette, 2008). Still, even with high ambitions, students

lack the fundamental knowledge of what is required to attain their ambitious aspirations. Not

knowing how much and what types of educational attainment are necessary to support their

31

occupational aspirations is problematic, as it leaves students unable to develop realistic plans for

their future. According to research on high school dropout and completion rates, only 73.4% of

high school students graduate with a high school diploma (NCES, 2009), even though 91% of

students expect to receive one (Yazzie-Mintz, 2009). Researchers have termed the inconsistency

in adolescents’ knowledge of the world of work and the educational pathways to fulfill their

occupational aspirations as an issue of misalignment (Csikszentmihalyi & Schneider, 2000;

Orfield, 1997; Rosenbaum, 2001; Schneider & Stevenson, 1999), noting that nearly 60% of

students have misaligned ambitions (Schneider & Stevenson, 1999). According to Paul (1997a),

misalignment between aspirations and preparation crosses social class, gender, race, and

ethnicity. Early theoretical work in vocational development (Ginzberg, Ginsberg, Axelrad, &

Herma, 1951) called the process of vocational development irreversible, since experience itself

and one’s subjective response to it shape vocational choices and build a foundation for future

vocational experiences. If the development process is misaligned, it can have negative

repercussions on aspirational development and attainment that can last a lifetime. Due to the

importance of student educational and occupational aspirations, the alarming rise in aspirational

misalignment, and the school’s role in it, becomes of equal import.

Level of student engagement can be an important indicator of misalignment. It is widely

accepted that the knowledge acquired through high school is related to later options for higher

education, and yet many students are disengaged from the academic work required of them

during their high school years (Csikszentmihalyi & Schneider, 2000; Rosenbaum, 2001).

Initially hypothesized by Stinchcombe in 1965, work bound students do not see school as

relevant to their future. This belief was also expressed by work and college bound students alike

in 1993 (Rosenbaum, 2001). Recent research on student engagement confirms that while more

32

students acknowledge that their efforts in high school matter in order for them to go on to

college, 46% of students are disengaged at school, believing that among other things, their work

is “pointless.” Two thirds of all respondents felt that school did not contribute to their growth in

areas such as thinking critically, understanding challenging material, or solving real world

problems (Yazzie-Mintz, 2009). Cognitive dissonance between effort and relevance to future

outcomes creates the backdrop for aspirational misalignment. If students do not make the

fundamental connection between their engagement in high school to subsequent high school

outcomes, consequential connections between their aspirations and the work required to attain

them may also be at risk.

Reasons for such large numbers of students with misaligned aspirations are numerous. One

factor in misalignment is structural. Educational policy which advances a “college for all” norm

(Rosenbaum, 2001, p. 56) may have good intentions such as increasing academic achievement

and closing the opportunity gap between students from differing SES backgrounds, but

according to many researchers the structure of high schools fails to meet the needs of college and

non-college bound students alike. First, a college for all norm ignores the needs of those

students who are a good match for stable jobs that do not require a college degree or post-

secondary skill sets (Reynolds et al., 2006; Rosenbaum, 2001). Counselors are not apt to

discourage any students from pursuing college and thus are less likely to suggest alternatives to

college (M. K. Johnson & Mortimer, 2002; Rosenbaum, Milller, & Krie, 1996) and the lack of

non-college guidance leaves students with few pathways for finding relevant information

regarding occupations and planning appropriately.

A similar problem holds true for students who desire to pursue a college degree. While

over 80% of students say they will pursue a college education (Goyette, 2008), only about half

33

will complete a college degree (Resnick & Wirt, 1996; Rosenbaum, 2001). Once again, the

students who do not complete a college degree may have had little educational preparation or

occupational guidance in high school to assist them when they are unsuccessful in their college

pursuits, even though their “subsequent failure is highly predictable in high school” (Rosenbaum,

1998; 2001, p. 58). Rosenbaum (1975, 1976, 1986) posited that high schools encourage high

educational aspirations but do not properly discuss barriers, what is necessary to overcome

barriers, or what type of planning and effort is necessary to successfully attain those educational

aspirations. For college bound students who go on to receive postsecondary degrees, the college

bound focus in high school may provide academic training, but a lack of information about

occupations in relation to their educational pursuits may still fail to assist students in aligning

their aspirations for future attainment. Schneider and Stevenson (1999) posited that these

students make less effective use of postsecondary resources than they would if they had a greater

sense of direction.

In addition to the college for all norm, unofficial policies within schools such as “social

promotion” contribute to the problem of student aspirational misalignment. While research has

long confirmed that grades are predictive of future success in college, students may not associate

grades and academic achievement with future opportunity because schools do not make this clear

to them. If students are promoted to the next grade level regardless of actual achievement, they

begin to confuse academic effort with time spent in the classroom. Given their experience, such

students are unlikely to expect college norms to be any different than those in high school, but

are likely to fail since this expectation is not reality. Social promotion in high school has

resulted in many students unknowingly being unprepared for college, with future aspirations

similarly misaligned.

34

The lack of structural support for school to job linkages has left students with little

knowledge of how to transition from school life to work life. The 1988 report by the Grant

Commission concluded that too many young people across the US flounder and fail in their

efforts to successfully make their way into adulthood. Glover and Marshall (1993) summarized

the US approach to structural school-to-work linkages by declaring that

America has the worst approach to school-to-work transition of any industrialized nation. Put simply, we have no systematic processes to assist high school graduates to move smoothly from school into employment. . . . Most high school graduates not going to college are left sink or swim--without advice or career counseling and without any job-placement assistance. (p. 130) Even with the School to Work Opportunities Act (1994), structural linkages between

schools and organizations encouraged by the policy failed to catch on, with most school districts

nationwide unsuccessful at seeking out and implementing such programs within their schools

(Orfield, 1997; Paul, 1997a). Although research acknowledges the need for structural linkages,

since the unsuccessful attempts with STWOA such an approach has not been widely pursued

(Csikszentmihalyi & Schneider, 2000; Rosenbaum, 2001; Stern, 2009). For all students, a high

school environment that fails to associate the importance and requirements of education with

relevance to future occupational success creates an environment ripe for aspirational

misalignment. In such an environment, students end up developing aspirations and aspirational

plans in isolation, unconnected to valid educational requisites or occupational knowledge

(Goyette, 2008). Glover and Marshall (1993) posited that these structural issues ultimately fail

the students the education policies were meant to assist. Interestingly, these structural issues also

signal that the aspirations of high schools themselves are as misaligned as a majority of its

students.

35

Given the lack of exposure to, and education about, linkages between school and work

and structural support for aspirational alignment, adolescents are left largely on their own to

make connections between school and work (Mortimer et al., 2002). Adolescents face greater

workplace complexity, with more occupational options and little knowledge about how to choose

between them or prepare for them (Modell, 1989). Without structural supports and knowledge

about what effort is required to meet their aspirational goals, M. K. Johnson and Mortimer

(2002) noted that students lack interest in thinking about their adult occupational choices.

Gottfredson (2005) suggested that the structure and protective nature of high school itself allows

students to delay making choices regarding their future and as such, students are less motivated

to obtain the information necessary to make informed career choices at a time when access to

resources, planning, and education are readily available. Students may also be unmotivated to

explore future occupational information because, as Csikszentmihalyi and Schneider (2000)

posited, adolescents often associate work with obligatory and undesirable activities. As such,

many adolescents view work only as palatable if it results in increased monetary reward, power

or fame. Lack of information and disconnection between academic effort and occupational

attainment leave these students with naïve and misaligned aspirations. The delay in serious

occupational consideration may be a prolongation of childhood with serious consequences in that

such students are left to grapple with reality after educational opportunities have already

narrowed and they are beyond the scope of secondary school guidance (M. K. Johnson &

Mortimer, 2002). Lack of exposure to the vast occupational landscape leaves many students

unable to plan appropriately and often, despite their level of academic achievement, they pursue

college “in the hope that their vocational goals will become clearer along the way”

(Csikszentmihalyi & Schneider, 2000, p. 215). Without the knowledge of whether or not college

36

may be needed for their ultimate aspirations (Rosenbaum, 2001), or even how to be successful

(Goyette, 2008) once enrolled, students take their unresolved problems of misalignment with

them from high school to college.

Misaligned aspirations establish a false foundation for future success of students. In an

educational environment in which students are asked to make choices but not given sufficient

information (Rosenbaum et al., 1997) and are promoted to subsequent grade levels and

encouraged to go on to college despite their academic achievement (Rosenbaum, 2001),

underachieving students may have an artificial sense that they are prepared for college and

develop aspirations and plans accordingly. Yet, research on the number of students with college

ambitions who actually receive a college degree confirms that often these students are not

prepared for postsecondary education. According to Rosenbaum (2001), open admission

policies at community colleges perpetuate inflated educational aspirations of low-achieving,

underprepared students with low barriers to entry and remedial course offerings. While students

who attend two-year community college programs often view them as a second chance to learn

what they missed in high school, research shows that only 18.7% of those planning to get a two-

year degree actually attain one in the ten years after leaving high school. Many students who

enter two year programs end up “downsizing” their occupational aspirations (Schneider &

Stevenson, 1999, p. 208). For those underprepared students who go on to four-year college

programs, only 44.5% succeed in getting that degree. Four-year colleges have also begun

offering remedial courses for students who are underprepared for regular college curriculum.

According to the National Center for Educational Statistics (1999), 40% of college students have

taken remedial courses. The effect of remediation is costly for students in multiple ways. First,

students pay college tuition for remedial courses as they would regular college courses but

37

receive no college credit for remedial work, which can be financially costly (Rosenbaum, 2001).

Second, research by Deil-Amen and Rosenbaum (2002) revealed that the more remedial courses

a student takes, the higher the probability that the student will drop out. This is emotionally

costly, affecting a student’s self-esteem (Rosenbaum, 2001), as few students pursuing college

expect to fail. It also leaves the student without a degree, limiting future earnings (S. R. Miller &

Rosenbaum, 1997; Rosenbaum, Miller, & Roy, 1996) and potentially limiting occupational

opportunities as well. Finally, for students who postpone occupational planning until college,

dropping out of college leaves them with little information, few skills, and no developmental

time to assess, plan, and align their aspirations. Data on misalignment confirms that students

who are unsuccessful in attaining their educational aspirations end up lowering their

occupational aspirations (Rosenbaum, 2001; Schneider & Stevenson, 1999).

The negative effects of misaligned aspirations are far reaching. Despite incentives, for

students who do not view their current high school experiences as relevant to their futures or

careers, they often become disengaged from school and post-school planning (Kenny et al.,

2003). Rather than focusing on consequences, research on student engagement found that

students tend to focus on their experiences within school (Yazzie-Mintz, 2009). In addition to

those who feel school is irrelevant to their futures, students at risk for disengagement include

students who feel a lack of importance, a lack of community, or a sense that someone at school

cares, and those who feel a lack of control over the system or their ability to change it.

Additionally, the research suggests connections between engagement and achievement and that

there is an engagement gap that mirrors the achievement gap in that the characteristics of

students with low levels of engagement are the same students with lower levels of achievement

(Yazzie-Mintz, 2009). Understanding school as relevant and being engaged with the work of

38

school is important for alignment. Without it, students are not only likely to have misaligned

aspirations but may drop out of school, which is the ultimate form of disengagement.

Ramifications for students who drop out include lower lifetime earnings and lower SES due to

limited occupational opportunities. Misalignment has another effect, which is the ratcheting

down of aspirational levels upon leaving high school. I. H. Lee and Rojewski (2009) showed

that career aspirations experience negative growth after high school graduation and a move to

less prestigious occupational goals.

The individual effects of misalignment have a larger, societal impact. While individual

disengagement may result in lower academic achievement, aspirations, attainment, and

individual earnings, it may also signal societal disengagement. On a large scale, disengagement

threatens societal goals and the ability to meet societal needs (AEE, 2009). The landmark report

on A Nation at Risk (National Commission on Excellence in Education, 1983) delineated the

societal need for the US to be academically prepared in order to be globally competitive. This

need has been reiterated in follow-up research (Weiss, 2003) and has been an important premise

underlying more recent educational policy mandates such as No Child Left Behind (No Child

Left Behind Act of 2001, 2002). Domestically, disengagement may result in higher

unemployment, incarceration rates, and reduced financial contribution (Sum, Khatiwada,

McLaughlin, & Palma, 2009), all with negative societal effects. As Rosenbaum (2001) noted,

the ability for a society to persist in its goals for another generation requires the efforts of

engaged youth, and disengagement from school may lead to disengagement from society,

threatening persistence. Massive misalignment may have similar effects. While misaligned

aspirations may result in disappointed educational and occupational attainment, it may also result

in more social stratification and less fluidity in the status attainment process. Educational

39

attainment is a primary determinant of occupational attainment, which influences the

consequential social status via economic attainment of individuals (Haller & Portes, 1973). As

noted previously by Blau and Duncan (1967), a less rigid social structure is necessary so that a

greater number of people gain the knowledge and skills once reserved for the elite as increased

technological needs will require greater numbers of trained and educated support people.

Society may be left without an adequately prepared workforce with the knowledge and skills

necessary for production. In this way, misalignment, disappointed educational aspirations, and

lack of occupational knowledge and linkages threaten societal progress. Misalignment has

serious ramifications for the educational system as well. While education policies reflect various

societal needs such as a desire for global competitiveness, the preparation of students for the

future, considerable misalignment decreases rather than increases viable educational and

occupational opportunities for students, calling into question the efficacy of the educational

system as a whole. Indeed, Paul (1997 a) summarized the problem of misaligned aspirations,

noting that “aspirations spur endeavor and persistence when they are connected to the programs,

paths and steps that can lead to positive outcomes. They are meaningless, even harmful, when

they are not connected to those programs, paths, and steps” (p. 133).

Importance of Locus of Control

Even where programs, paths, and steps exist, individual agency is a critical factor in

achieving outcomes (Seifert, 2004), as structural and agentic processes work in tandem

(Mortimer, Staff, & Lee, 2005). According to the research by Csikszentmihalyi and Schneider

(2000), two factors that shape the transition from youth to adult careers are social forces and

personal effort, and that “individual agency is no less real and important to career formation than

40

external circumstances” (p. 57). Foundational work in the area of vocational psychology by

Super (1953) noted that self-evaluations play an important role in influencing the development of

aspirations. Schoon’s research (2001) found that career development is mediated by both

individual and contextual factors, and in his early research, Kerckhoff (1972) found that social

psychological dimensions such as self-concept and sense of control over one’s environment

mediate relationships between outcomes and other individual variables such as SES.

Examination of status attainment and educational and occupational attainment by Wang et al.,

(1999) using multivariate regression on 1,927 respondents in the NLS 72-79 found that after

accounting for background variables, locus of control (i.e., sense of control over one’s life

choices) has significant effects on the attainment for both education (.135) and occupation (.097)

and the effects were greater than those of self-esteem. This finding has been called into question

by Cebi (2007) who examined the effect of locus of control on educational outcomes and

occupational expectations of 1,737 high school students in the National Longitudinal Survey of

Youth and found that when controlling for background variables, locus of control does predict

educational attainment and the likelihood of attending college. However, once the level of

cognitive ability is introduced, locus of control ceases to be significant. Cebi’s work noted that

locus of control is rewarded later in the form of higher wages. Still, given the importance of

agency to occupational formation and attainment, understanding students’ perceptions of seminal

constructs such as locus of control and how much control they feel over their life choices or their

environment is also important.

Additional research suggests that awareness of barriers related to factors such as SES,

gender, and academic attainment have direct and negative effects in limiting aspirations of youth

(Gottfredson, 1981; Kenny et al., 2003). Indeed, perceived barriers may have as much influence

41

on career attitudes and behaviors as actual barriers (Flowers et al., 2003; Swanson & Woitke,

1997) and in work by Kenny et. al., (2003) perceived barriers emerged as a negative predictor of

educational and vocational attitudes. Mortimer et al. (2005) found similar results confirming that

“agency-related psychological precursors have significant effects on achievement-related

behaviors” which are associated with attainment even after accounting for SES (p. 144). As

Mortimer et al. noted, “persons subjectively orient themselves to the opportunities and

constraints that they encounter in particular times and places, and the strategic actions they

devise to reach their goals in particular structural contexts” (p. 133). This is echoed in work by

Smith-Maddox (1999) who offered that aspirations and outcomes are influenced by the social

and cultural resources embedded in students’ environments, such as social class or curriculum

track. Rosenbaum (2001) suggested that student efforts are based not solely on internal

motivation (which may be influenced by environment) but by their perception of the future

relevance (outcome) of what they are doing, perceptions which previously cited research shows

may be influenced by social psychological factors.

The importance of the interactions between contextual environment, personal

perceptions, and subsequent action aligns with social learning theory (Rotter, 1954), which

specifically examines human behavior as influenced by environment, cognition, and behavior

(Bandura, 1977). Rotter (1954) believed that “major or basic modes of behaving are learned in

social situations and are inextricably fused with needs requiring for their satisfaction the

mediation of other persons” (p. 84), thus describing his theory as social learning. Social learning

theory has two main concepts which impact behavior: reinforcement value and expectancy.

Briefly, reinforcement value is “the degree of preference for any reinforcement to occur if the

possibilities of their occurring were equal,” while expectancy is “the probability held by the

42

individual that a particular reinforcement will occur as a function of a specific behavior on his

part in a specific situation or situations” (Rotter, 1954, p. 107). In other words, people have

preferences for outcomes that impact their behaviors, which are learned from prior experience or

expectation of a situation.

Extending his original theory, Rotter (1966) proposed a construct which he called locus

of control of reinforcement in which he theorized that people may or may not believe they can

control the reinforcements which will in turn impact their behavior. Subsequent reinforcements

over time strengthen or weaken behavioral responses. Rotter identified two loci of control of

reinforcement: internal and external. Having an internal locus of control (internals) refers to

attributing a reinforcement (outcome) to one’s own behavior; feeling that one has control over

the outcome and behaving accordingly. Possessing an external locus of control (externals) refers

to attributing an outcome to something outside of one’s control, such as luck, fate, other people,

the powers that be, unpredictability, or turbulence. Externals feel that outcomes have little to do

with one’s behaviors and thus those behaviors are less likely to be influenced (strengthened or

weakened) by outcomes. One variable that influences the size of expectancy is the number of

previous experiences with a particular situation; the more experiences, the larger the expectancy

value. The fewer experiences, the less the experience will affect expectancies and thus a smaller

expectancy value. However, with fewer experiences, whatever expectancies there are will most

likely reflect the last similar experience (Phares, 1976). Locus of control is accepted as a

relatively enduring dispositional characteristic which becomes modified through experience

(Findley & Cooper, 1983). As noted by Mirowsky and Ross (2003) and Schieman and Plickert

(2008), perceived control represents a cognitive link between objective, social conditions, and

inner experience.

43

Commonly referred to as a motivational variable (deCharms, 1968) and one of the

dimensions of motivational attributions (Graham, 1991; Seifert, 2004; Weiner, 1984), research

on locus of control spans multiple disciplines, has shown Rotter’s (1966) locus of control

construct to be robust, and reveals some interesting findings on its importance to educational and

occupational aspirations and attainment. Indeed, students pursue both academic and social goals

in the classroom (A. M. Ryan & Patrick, 2001) and, according to Ames (1992), the perception of

control appears to be a significant factor affecting student engagement in learning as well as the

quality of learning. In fact, J. S. Coleman et al. (1966) found that locus of control was not only

an important predictor of academic performance but the most important determinant of

educational achievement than any other factor in a student’s background (M. Coleman &

DeLeire, 2003). Stipek (1980) examined locus of control and achievement with 89 middle and

lower SES students in first grade using both panel correlation and path analysis and found that

internal locus of control impacts student achievement. Ross and Broh (2000) analyzed a sample

of 8,802 students from the base year, first and second follow ups of the National Educational

Longitudinal Study (NELS:88) and using a confirmatory factor analysis found that academic

achievement in the 8th grade is associated with an internal locus of control in 10th grade which in

turn is associated with higher academic achievement in the 12th grade. Additionally, Ross and

Broh confirmed that self-esteem has a smaller impact on academic achievement and does not

improve grades or test scores. Using the same dataset with a different sample of 18,311 students

spanning the base year through the second follow up, Rojewski and Yang (1997) analyzed the

influence of specific factors on occupational aspirations using structure equation modeling and

found academic achievement and self-evaluation (specifically locus of control and self-esteem)

had “consistent, positive, and statistically significant influences on occupational aspirations,”

44

(p 403) even though the effect size was moderate. Interestingly, their study found that these

effects decreased over time from 8th to 12th grade. Still, Murasko (2007) found that locus of

control may have latent and cumulative effects on educational and occupational aspirations and

outcomes.

Externals

Based on Rotter’s model and scale for measuring locus of control, much of the research

on the construct focuses on the impact of the control orientation. Individuals with an external

locus of control orientation (externals) often believe that fate, luck or other people control the

reinforcements that influence their behavior and outcomes. Externals do not often feel

responsible for their lives and are less likely to trust their own abilities or persist through difficult

situations. External locus of control is often associated negatively with outcomes. People with a

more external locus of control are often found to be less motivated, lower achieving and having

lower aspirations and attainment than those with an internal locus of control. Externals are likely

to attribute failure to uncontrollable factors, and are unlikely to experience pride, satisfaction,

confidence, or self-esteem (Seifert, 2004). Rumberger (Rumberger & Palardy, 2004) found that

students with a high internal locus of control orientation were less likely to drop out of school

while research in the Journal of College Admission found that students with an external control

orientation were more at risk for dropping out of college (Gifford, Mianzo, & Briceno-Perriott,

2006). Indeed, one’s locus of control orientation has an effect on aspirations and outcomes.

Thus, even when choice making is constrained by actual or perceived barriers, people continue to

make choices, but may feel that they are not completely in control of making those choices

(Mortimer et al., 2002). Research on students with learning disabilities often cites external locus

of control caused by repeated failure and inability to act as a self-advocate as problems for such

45

students (Rojewski, 1996). Research by Glasgow, Dornbusch, Troyer, Steinberg, and Ritter

(1997) found that students with non-authoritative parents were more likely to have an external

locus of control

Internals

Individuals with an internal locus of control orientation (internals) are more likely to

attribute their behavior and outcomes to internal, controllable factors. An internal locus of

control orientation has been found to be effective in motivating persistence and achievement.

Relationships between locus of control and achievement found in the literature suggest that

students with a higher sense of work and a greater sense of control over their outcomes are more

productive at school and have better outcomes (Murasko, 2007), and according to attributional

research by Weiner (1984) students who attribute success and failure to internal, controllable

causes are more likely to feel pride, satisfaction, and confidence and are more likely to “choose

to work on more difficult tasks, persist longer in the face of failure and display higher levels of

cognitive engagement and produce work that is of higher quality” (Seifert, 2004, p. 140).

Weiner (1984) also created a distinction within the internal locus of control orientation in that

students may attribute failure to internal but uncontrollable factors (inability) and in that case are

more likely to feel shame and humiliation and will show little effort or cognitive engagement

(Seifert, 2004). Seifert (1997) described individuals in pursuit of mastery goals to be typically

self-regulating and self-determining individuals that attribute their success or failure to internal,

controllable factors. The disposition of such individuals fosters cognitive development.

Csikszentmihalyi and Schneider (2000) noted that in addition to aspirations and information,

students must develop positive personal traits and attitudes such as an internal locus of control to

help ensure a smooth transition between school and the adult working world—although they also

46

point out that feeling one can control all events through hard work and determination despite real

external barriers can have, among other things, detrimental health effects (p. 220). Research by

Hanson (1994) found that low levels of internal locus of control had significant and negative

effects on aspirational development, while Mau and Bikos (2000) found that an internal locus of

control was a positive and significant predictor for both educational and occupational aspirations.

Early research on student locus of control commonly found that students from middle or higher

SES family backgrounds have, on average, a more internal locus of control orientation than

students from lower SES families (Bartel, 1971; Battle & Rotter, 1963; Shaw & Uhl, 1971;

Stipek, 1980).

Research confirms that motivations and attitudes are important despite all plans: doing

well in high school and college lead to increased opportunities in the adult world of work (Inoue,

1999). Work on the motivational variable locus of control confirms that specific attitudes about

one’s control over one’s work may significantly affect the quality of that work, aspirations, and

future attainment. With clear connections between locus of control and aspirations, it becomes

important to understand how the context of school may impact student locus of control.

Schools Impact on Student Locus of Control

Schoon (2001) observed that the transition from school to work is shaped by contextual

forces which in turn are acted upon by the individual (p. 124). According to Csikszentmihalyi

and Schneider (2000), schools are the main arenas in which children prepare themselves to take

on the responsibilities of the adult world and are the main institution for socialization before

adulthood. In their work on schools, academic motivation and stage-environment fit, Eccles and

Roeser (2009) noted that “understanding the impact of schools on adolescent development

47

requires a conceptual framework for thinking simultaneously about schools as contexts in which

development takes place and about the changing developmental needs of students as they move

through the schools system” (p. 404). Eccles and Roeser analyzed the context of schools as

systems with multiple levels and noted that the multi-level process are dynamic, highly

interdependent, and may be complementary or contradictory and may influence students either

directly or indirectly. Earlier work argued that while students are developing as they transition,

the whole nature of schools is also changing and that these changes can affect the motivation and

behavior of adolescents (Eccles & Midgley, 1989). These changes may impact motivational

variables such as locus of control either positively or negatively.

One area identified by Eccles and Roeser (2009) where schools impact student

motivation is the process of academic sorting and tracking. Researchers on stratification in the

US, note that the grouping of students between and within schools has significant effects on later

levels of academic success (Alexander & Entwisle, 2000; Gamoran, 2000) which in turn affects

occupational outcomes (Cook et al., 1996). Kerckhoff (1995, 2000a) pointed out that the sorting

process itself impacts academic achievement and even if offering the same generic educational

credentials, the sorting mechanisms alter the distributions of probabilities of obtaining them. He

argued that changes in educational status ambitions reflected students’ responses to status

classifications and resources allocations experienced within school, stating that the sorting

decisions made by teachers and counselors provide students with information about themselves

and their probable future and “create socially significant classifications on the basis of which

others will respond to them differentially” (Kerckhoff , 1976, pp. 374-375). Similar work by

Rosenbaum (1976) noted that the information transmitted each school day from kindergarten on

up causes student to negotiate their identities, which drive the decision process, including

48

aspirational development (Paul, 1997b). Those responses help motivate students in higher ability

groups and alienate students in lower ability groups (Gamoran, 2000). In other words, students

placed in consistently higher or lower ability groups in areas such as reading or math may begin

to view this placement as a form of reinforcement of their performance, subsequently affecting

their motivation. Students may also begin to identify their ability status with their understanding

of themselves (identity).

As noted previously, research shows that sorting and ability grouping affect students in

part because of changes in teaching and instruction methods related to those ability groups

(Gamoran, 1993; Gamoran & Berends, 1987; Slavin, 1990). While students may not recognize

the meaning of sorting within schools at early ages, teachers and counselors may have differing

expectations of students based on sorting and thus make recommendations and decisions about

future placement that may have long-term consequences (Gamoran, 2000; Paul, 1997a). For

students who perceive the sorting process and differences in outcomes and opportunities related

to the types of schools and classrooms, school itself may be recognized as a form of

environmental barrier to aspirational attainment. Kerckhoff (1993) found that there was a level

of consistency in how students were sorted throughout their educational careers and into the

labor force which could not be attributed only to differences in skill level but to the long-term

allocation itself. Such long-term allocation to a specific group and indirect effects of

socialization work in concert and have cumulative effects on how students feel about themselves

and how committed they are to school and successful academic work which in turn affects

students’ life chances. More specifically, students’ recognition of barriers may change their

outcome expectancy, especially if such barriers have been reinforced over time. By the end of

high school, their level of motivation (Eccles & Midgley, 1989) and aspirations may have

49

decreased significantly (Helwig, 2004). In sum, the perception that schools control a student’s

future may impact their locus of control.

While schools may present structural barriers to some student aspirations via ability

grouping and recommendations, they may also assist students in learning to navigate around

other structural barriers they may find when they enter the labor market. While students may

feel little control over structural barriers, schools are important in that they may help students

increase their internal locus of control through guidance counseling assistance in identifying

sources of job training, helping students with applications materials, job searching strategies, and

encouraging the accumulation of meaningful skills and credentials, including high school and

postsecondary education (M. K. Johnson & Mortimer, 2002, p. 69). Early research by Bradley

and Gaa (1977) revealed that schools can increase student internal locus of control through goal

setting projects. Subsequent research on the positive effects of counseling on a specific group of

students who were in seventh and eighth grade and whose parents had divorced found that group

counseling with specific activities such as role playing and peer discussions increased those

students’ internal locus of control (Omizo & Omizo, 1988; Whiston & Sexton, 1998). Research

by Plucker (1998) examined the relationship between school climate conditions and student

aspirations by analyzing a sample for 1,170 students from the Secondary School Aspirations

Survey and found that specific school climate variables such as academic press, mentoring, and

community are associated with higher levels of self-confidence (including locus of control items)

and aspirations. Stipek (1980) proposed that schools may foster the development of internal

locus of control through task-oriented work that allows students to observe the success or failure

of their efforts and through directly teaching student’s responsibility for their work and behavior.

Similar to work toward educational goals through locus of control, research by vocational

50

psychologists proposes that occupational development also requires the development of an

internal locus of control as early as elementary school (L. S. Johnson, 2000; Schultheiss, 2005;

Super, 1953, 1980), since occupational prestige levels are established early in life and are

resistant to later intervention (Gottfredson, 1981). Work by Turner (2007) in which she used

structure equation modeling to analyze a sample of 147 inner city eighth graders identified

specific factors which schools can actively foster that promote persistence in educational and

vocational development, including: academic preparation, career development skills training,

parental assistance, and skills to overcome social and environmental barriers. Similar research

by R. Ryan and Deci (2000) and Seifert and O’Keefe (2001) suggested that fostering competence

and control is critical since they influence behavior, and propose that the interaction between

teachers and students is a key factor in such development. Specifically, teachers perceived as

nurturing and supporting help students develop confidence and self-determination which

translate into the types of behaviors associated with internal motivation.

School impact on student locus of control is important in that locus of control is linked to

the outcomes such as engagement, quality of learning and academic achievement. In terms of

achievement, ongoing research establishes that achievement and perceptions of control are

related (Bartel, 1971; J. S. Coleman et al., 1966; Graham, 1991; Mau & Bikos, 2000; Murasko,

2007; Stipek, 1980) and reinforce each other (Bradley & Gaa, 1977; Ross & Broh, 2000; Stipek,

1980). J. S. Coleman et al.’s (1966) landmark educational research reported that students’ sense

of control was a better predictor of achievement than any of the other school and family variables

measured. Indeed, schools that foster environments geared toward both achievement and support

of social psychological attributes such as locus of control (which has specifically been linked to

enhancing achievement) will enhance the locus of control and aspirations of its students (Neild,

51

2009). Both school organization and climate (Rumberger, 1995) are important for fostering

achievement and internal locus of control.

Importance of Study

There is general agreement that positive student outcomes are the desired goal of

education. Yet, with disturbing proportions of students with misaligned aspirations and

subsequently negative outcomes, understanding specifically how schools can shift this negative

trend is paramount. As noted in the research literature, an important variable in the development

of student aspirations is locus of control. Although research on the role of schools in aspirational

development has a long history, and how schools may impact student locus of control has also

received attention, the examination of whether and how schools specifically impact student

aspirations via their impact on student locus of control would contribute an important piece to

the research. According to Kerckhoff (2000 b), “There has been little attention to differences in

culture, social structure or the institutional linkages between school and work that could lead

certain psychological dimensions to be more important for achievement in one context than

another” (p. 26). Kerckhoff holds that such studies could yield “understanding of the

interactions of structures and attitudes in predicting attainment outcomes” (p.26). Indeed, there

is a need to understand the extent to which personal control impacts aspiration development and

more specifically, the impact of the school environment on a students’ sense of control and how

that locus of control then impacts occupational aspirations—since such aspirations have been

shown to have significant effects on future actions regarding attainment. As noted by Cebi

(2007), locus of control is potentially important in analyzing the investment schools make in

children (p. 931), and may hold answers to how schools reverse the trend in misalignment.

52

CHAPTER III

RESEARCH MODEL

Theoretical Framework and Conceptual Model

Rotter’s work on social learning theory (1954) and his subsequent theory on locus of

control of reinforcement (1966) generated a large body of robust research around this social

psychological construct. As previously noted, Rotter’s social learning theory posited that there

exists an interaction between people and their environments that motivates behavior. Since the

main focus of this study is whether environmental characteristics of schools impact the

development of student occupational aspirations over time, social learning theory serves as the

primary frame for this study’s model in which student interaction with schools environments

may affect their motivation and subsequent occupational aspirations.

While the main dependent variable in this study is occupational aspirations, a key interest

is whether student locus of control is a potential mediator through which school contextual

characteristics might influence student occupational aspirations. Since a thorough review of the

literature reveals that both academic track and academic achievement have been shown to

influence the formation of student occupational aspirations and may also relate to student locus

of control, consideration of these variables is also important. In addition, the development of

student occupational aspirations happens over time, and as such, consideration of the process of

development over time in relation to the individual student is equally important.

By definition, a discussion of process assumes a sequence of events, with both a

beginning and an end. However, in a dynamic process related to aspects of human development

53

(i.e., locus of control), the start and stop points of a process, or specific influences at specific

points in time, may be difficult to identify. Additionally, processes may be happening

simultaneously (i.e., taking classes while making decisions about one’s aspirations) and over

time, the effects of such processes may be cumulative, as noted by Rotter’s social learning theory

(1966). Given these complexities, what follows is a theoretical discussion of the dynamics of the

development of the core dependent variable (student occupational aspirations over time), while

considering the influences of key variables noted in the literature review, including student

academic track, academic achievement, and locus of control in relation to specific school

contextual variables (counseling support, school climate, academic press, school SES). While a

more detailed description of the specific variables to be included in the study is described in

Chapter IV, the following is an overview of the main elements in relation to the dynamic

framework posited by Rotter’s social learning theory.

Occupational Aspirations

Although students are exposed to occupational discussions early on in their academic

careers (i.e., career days in elementary school), the links between occupations and self are often

forged later, during middle school and junior high years (Helwig, 2004; I. H. Lee & Rojewski,

2009). During this time, students begin connecting occupations with personal endeavor. With

schools being a primary source of information for students, the availability of school resources

and support available may affect whether and how student make connections between

occupations and achievement. Indeed, if resources are scarce, unavailable or unknown to

students, such connections may be incomplete or misaligned and the potential ramifications may

not be well understood by the students. If, for example, teachers do not discuss occupations in

54

relation to schoolwork, encourage students to speak with counselors about occupations and

explicit educational requirements or schools do not have counseling resources available to

students, linkages between academic effort and occupational outcomes may remain fuzzy to

students, and their occupational aspirations may be misaligned. Of course, one would expect

that more focused interaction between students, teachers and counselors around occupations and

academic requirements would result in students developing more aligned occupational

aspirations. Still, even with such expectations, additional considerations in the development of

student occupational aspirations are changes in student context (i.e., changing to a different

school in a different neighborhood or city or town) as well as the cumulative effects of variables

such as curriculum track or academic achievement from previous years.

Locus of Control

An initial point of consideration is that of the student’s own sense of personal control.

Research on locus of control outlines an external and internal locus of control, with external

locus identified as one feeling little or no control over future events and an internal locus

identified as one feeling a personal sense of control over the future. In itself, this locus of control

(which is itself influenced by SES (Bartel, 1971; Battle & Rotter, 1963; Shaw & Uhl, 1971;

Stipek, 1980).) might inform occupational aspirations without any other variables. For example,

a student with a highly internal sense of control might aspire to be a doctor, while a student with

a highly external sense of control might aspire to occupations with lower level of prestige since

they do not feel that they have any control over the final outcome and do not want to be

disappointed if they were to aspire to something higher and that aspiration did not come to

fruition. Still, it seems more plausible that the sense of control a child develops is, as noted by

55

Rotter (1966), gained through repeated experiences, such as those found in the place the child

spends a great deal of time: school.

Academic Tracking, Effort, and Achievement

The primary activity of school is for students to gain critical knowledge in the classroom,

and schools provide students with constant assessment as to how well they are doing at this

activity. Such assessment provides the child with experiences upon which to develop a locus of

control. Cumulatively, these assessments form the foundation for future educational opportunity.

If a child is doing poorly in school, their educational opportunities may be more limited than

those for the student who is doing well. According to the research, students very early on gain a

sense of where they stand in the hierarchy of assessment, often through early curriculum tracking

via assignment to a specific reading group or math class. Research posits that early curriculum

tracking often determines subsequent curriculum tracks (Alexander & Entwisle, 2000; M. K.

Johnson & Mortimer, 2002), and while there is a sense that students can choose which curricular

track they desire once reaching high school, students and parents often depend upon middle

school counselors to make the best recommendations for the child entering high school, and once

those recommendations are made, few students switch tracks in the course of their high school

careers (Paul, 1997a). While research asserts that the curricular track rarely changes in the

course of high school, even if there were a change in the track, it is theorized that the original

track would continue to impact student locus of control. And while parents are depending on

middle school counselors for curriculum recommendations, those counselors often depend on

previous assessments of students to make those recommendations. Thus, it is theorized that

assessments of academic achievement and subsequent curriculum tracking at individual points in

56

time and cumulatively over time may in turn foster development of either an internal or more

external student locus of control.

Additionally, student locus of control may determine the level of effort they put into their

academic pursuits. If an external locus, the effort may be low; if internal, the effort may be high.

The effort has subsequent ramifications in that it is reflected in the student’s academic

achievement. Being graded on academic effort, positively or negatively, may reinforce their

sense of control in subsequent pursuits, and in tandem may impact the way students think about

their future occupational capabilities and thus, the occupations to which they aspire.

Diagram of Theoretical Model

Given the theory, at an early stage in which students are beginning to consider self in

relation to occupation, an initial model might look like Figure 1, in which curriculum track may

be influencing student locus of control, which in turn may be influencing both academic effort

and occupational aspirations. The amount of academic effort, in turn, may affect student

Figure 1. Theoretical Model During Early Stage of Identification Between Self and Occupation (i.e., Middle School).

57

academic achievement, which may then impact occupational aspiration, since according to Lent

(1996) students modify their aspirations based on learning experiences and performance

outcomes. The variables on the left side are school contextual variables, which are only

discussed briefly here, but are significant to the whole of the theoretical model itself. These

variables may influence both student locus of control and may have an impact on student effort

and academic achievement.

Then, considering student progression through their academic career, one must also

consider how the variables in the model vary over time. For example, research by Ross and

Broh (2000) confirmed that academic achievement in 8th grade impacts locus of control in 10th

grade which impacts subsequent academic achievement in 12th grade. Additional research in the

area confirms the general understanding that previous locus of control impacts future

performance. One might theorize this as a type of dynamic feedback loop, with locus of control

acting on academic achievement via student effort. The extension of this theory is the impact of

academic achievement on occupational aspirations. Thus, the development of aspirations may be

impacted by locus of control both independently and via the effect of locus of control on effort

and subsequently, academic achievement. This process continues as a student progresses

through consecutive grades, assimilating information about their performance, and interpreting

their ability to control outcomes related to their efforts, with cumulative effects (Murasko, 2007).

An extension of the theoretical model in Figure 1 would show the continued and

cumulative impact of variables working as a type of feedback loop in which students are in

various and constant stages of assessing experiences and developing a locus of control, exerting

academic effort, having their performance assessed, assimilating their school experiences (i.e.,

school climate, academic press) and performance assessments into their locus of control, and

58

making sense of how their performance and locus of control might relate to their future

occupations. While the process of academic effort and assessment of academic achievement

happens sequentially, the model as a whole is happening all the time, and at different times for

each student. In the following extended theoretical diagram, the school contextual variables to

the left have a continued effect, even though they are only represented once. There are three

time points noted as a way to illustrate the dynamics of the variables over time.

One key difference in the diagram of the theoretical dynamic model noted in Figure 2 is

that it takes into account how student occupational aspirations may also impact student academic

effort and thus impact academic achievement. While students in earlier stages of their academic

careers may have less concrete aspirations (Kerckhoff, 1974; Rindfuss, et al, 1999) and as such

may not work hard academically with those aspirations in mind, students in later stages of their

academic careers may be more likely to consider their aspirations when exerting effort in their

Figure 2. Dynamic Theoretical Model Illustrating the Effects of Tracking and Academic

Achievement on Locus of Control and Occupational Aspirations Over Time.

59

academic pursuits. Another difference is that academic achievement continues to vary and have

varying effects on locus of control and occupational aspirations, where the curriculum track

becomes more invariant over time.

Diagram of Final Empirical Model and Implications

Since curriculum track becomes more invariant over time (Paul, 1997b), and since

student achievement provides information regarding the outcomes of student effort (Carbonaro,

2005) and has stronger ties in the literature to locus of control and aspirations than the student

effort variable, a more parsimonious model will not include student effort but will include

curriculum track as a control variable at the student level rather than as a predictive variable.

Thus the final empirical model will focus more succinctly on the impact of school contextual

variables on occupational aspirations via locus of control. In addition, the study will include an

analysis of the relationship between locus of control and academic achievement, first with an

exploration of whether locus of control might partially mediate the relationship between school

characteristics and academic achievement, and second with an exploration of whether locus of

control might, in addition to its hypothesized direct effects on occupational aspirations, may also

have indirect effects on occupational aspirations via academic achievement since this additional

analysis may provide an alternative understanding of the role of locus of control.

Figure 3 provides an overview of the relationships hypothesized and examined in this

study. As noted in the model, the relationship between school context and growth in student

occupational aspirations is proposed to be fully mediated by growth in student locus of control

and growth in student academic achievement. As such, school context is regarded as having a

direct effect on growth in student locus of control and growth in student academic achievement,

60

Figure 3. Dynamic Model of Full Mediation Between School Context Variables and Occupational Aspirations via Locus of Control and Academic Achievement.

which in turn directly influence growth in student occupational aspirations. Thus, with full

mediation (as depicted by the dotted line), when the effects of locus of control and academic

achievement are accounted for, any direct link between school context and occupational

aspirations is expected to become nonsignificant (Baron & Kenny, 1986).

This study also examines the relationship between locus of control and academic

achievement in which it is proposed that locus of control will have a direct effect on academic

achievement and an additional indirect effect on occupational aspirations via academic

achievement. Given this, there would also be an additional path for school context to indirectly

influence occupational aspirations via its relationship with locus of control.

If the analysis reveals that locus of control has a direct effect on academic achievement,

and academic achievement has a direct effect on occupational aspirations, the total effect

(indirect and direct effects together) of locus of control on occupational aspirations is

strengthened.

61

Finally, there is the potential for locus of control to mediate school context and academic

achievement. Given the strength of the literature outlining the relationship between school

context and academic achievement, it is not expected that there would be complete mediation,

but locus of control might partially mediate that relationship, which means that with the

inclusion of locus of control, the direct effects of school context on academic achievement would

be smaller. This relationship is hypothesized and will be modeled and analyzed.

Research Question and Hypotheses

Research Question

Do student locus of control, student occupational aspirations, and student academic

achievement change over time? (Findley & Cooper, 1983; Gottfredson, 1981)

Hypotheses

The following hypotheses are proposed. The variables noted in these hypotheses will be

defined and discussed in detail in Chapter IV.

Hypothesis 1: School context and occupational aspirations. Positive school

contextual characteristics (positive climate, counseling support, at or above average school SES,

and higher academic press) will be associated with gains in occupational aspirations over time

(Alwin & Otto, 1977; Caravello, 1958; Glover & Marshall, 1993; Helwig, 2004; Hughes &

Karp, 2004; Kelly & Lee, 2001; Kerckhoff, 2000a; McWhirter et al., 2000; Mortimer, 2000;

Orfield, 1997; Paul, 1997b; Plucker, 1998; Rosenbaum et al., 1997; Rothney, 1958; Schoon,

2001).

62

Hypothesis 2: Student locus of control and occupational aspirations.

2a: Positive school contextual characteristics (positive climate, counseling support, at or

above average school SES and higher academic press) will be associated with change toward an

internal locus of control over time (Bradley & Gaa, 1977; Csikszentmihalyi & Schneider, 2000;

Eccles & Midgley, 1989; Eccles & Roeser, 2009; Flowers et al., 2003; M. K. Johnson &

Mortimer, 2002; Kenny et al., 2003; Rosenbaum, 1976; Rotter, 1954; R. Ryan & Deci, 2000;

Schoon, 2001; Seifert & O'Keefe, 2001; Smith-Maddox, 1999; Stipek, 1980; Swanson &

Woitke, 1997).

2b: The more internal a student’s locus of control over time, the more likely the student is

to have gains in occupational aspirations over time (Cebi, 2007; Csikszentmihalyi & Schneider,

2000; L. S. Johnson, 2000; Mau & Bikos, 2000; Mortimer et al., 2005; Rojewski & Yang, 1997;

Schultheiss, 2005; Super, 1953; Wang et al., 1999).

2c: Student locus of control partially mediates the relationship between school contextual

characteristics (positive climate, counseling support, at or above average school SES, and higher

academic press) and occupational aspirations (Cebi, 2007; Helwig, 2004; Kerckhoff, 1972,

2000a; Paul, 1997b; Schoon, 2001).

Hypothesis 3: Student academic achievement and occupational aspirations.

3a: Positive school contextual characteristics (positive climate, counseling support, at or

above average school SES, and higher academic press) will be associated with gains in academic

achievement over time (Alexander & Entwisle, 2000; Borders & Drury, 1992; Eccles &

Midgley, 1989; Flowers et al., 2003; Franklin, 1995; Gamoran, 2000; Kenny et al., 2003;

Kerckhoff, 1993; V. E. Lee & Bryk, 1989; V. E. Lee et al., 1993; Orfield, 1997; Rothney, 1958;

A. M. Ryan & Patrick, 2001).

63

3b: Gains in academic achievement over time will be associated with increases in

occupational aspirations over time (Borders & Drury, 1992; Fouad, 1995; Gerler, 1985;

Kerckhoff, 1993; Lapan, 2004; Lapan et al., 2007; Lapan et al., 1993; Lapan et al., 1997; Mau,

1995; Orfield, 1997; Rosenbaum, 2001; Schneider & Stevenson, 1999; Whiston & Sexton,

1998).

3c: Student academic achievement partially mediates the relationship between school

contextual characteristics (positive climate, counseling support, at or above average school SES

and higher academic press) and occupational aspirations (Borders & Drury, 1992; Cook et al.,

1996; Lapan, 2004; Lapan et al., 2007; Lapan et al., 1993; Lapan et al., 1997; Paul, 1997b;

Whiston & Sexton, 1998).

Hypothesis 4: Locus of control and academic achievement.

4a: The more internal a student’s locus of control, the more likely the student is to have

gains in academic achievement over time (Bradley & Gaa, 1977; Cebi, 2007; J. S. Coleman et

al., 1966; M. Coleman & DeLeire, 2003; Graham, 1991; Hanson, 1994; Mau & Bikos, 2000;

Mortimer et al., 2005; Murasko, 2007; Ross & Broh, 2000; Seifert, 2004; Wang et al., 1999).

4b: Student locus of control will mediate the relationship between school contextual

characteristics (positive climate, counseling support, at or above average school SES, and higher

academic press) and student academic achievement (Bartel, 1971; J. S. Coleman et al., 1966;

Kerckhoff, 2000a; Mau & Bikos, 2000; Neild, 2009; Rumberger, 1995; Stipek, 1980).

64

CHAPTER IV

DATA AND METHODS

Dataset/Sample: NELS:88

In order to test whether the school context variables noted in the theoretical model have

an effect on student locus of control and occupational aspirations over time, a longitudinal

dataset is required that contains data on both students and the schools they attend. The National

Educational Longitudinal Survey (NELS:88) (Berkner, 2000; NCES 1996a, 1996b, 2000) is an

appropriate dataset for this study because it is a large and nationally representative study of 8th

grade students who go on to attend high school and contains repeated measures of students and

schools. NELS:88 is also important because it begins surveying students in 8th grade, which

provides a baseline prior to starting high school. Beginning with eighth graders in 1988 (base

year), NELS followed up with students every two years from 8th grade through the age of 20

with a final follow up six years later at the age of 26. Since two follow-ups were conducted

during the 10th (1990--1st follow up) and 12th grade (1992--2nd follow up) high school years, it is

possible to measure change between baseline locus of control, achievement and occupational

aspirations in 8th grade and locus of control, achievement, and occupational aspirations in 10th

grade and again at 12th grade.

In 1988, the NELS began surveying over 24,000 students in the 8th grade from 1,052

schools (815 public and 237 private), with the intent to “study the educational, vocational, and

personal development of students at various stages in their educational careers, and the personal,

familial, social, institutional, and cultural factors that may affect that development” (Curtin,

Ingels, Wu, & Heuer, 2002, p. 2). The study employed a two-stage stratified sampling method

65

with schools as the first stage and a random sample of students within those schools as the

second stage. The study ensured that it garnered a representative sample of students by

oversampling schools with larger populations of minority students, increasing the number of

randomly selected students in those schools by an average of two students. Overall, an average

of 25 students was selected from each school. For schools with less than 25 students in 8th grade,

all eligible students were selected (Ingels et al., 1994). Over 90% of the students completing

surveys in the base year also completed follow-up questionnaires in their sophomore and senior

years. Since this study examines the effects of school characteristics on students, it may be

important to note that while the sample of students is random, the sample of high schools is not,

since the schools are chosen based on where the sample of 8th grade respondents ended up

attending high school. To ensure against bias due to this lack of random sampling of high

schools, the models in the study must be properly specified (Gamoran, 1996), which is discussed

in greater detail in the section which outlines the models. For this study, only students who were

in the appropriate grade during the base year (8th grade), first (10th grade), and second (12th

grade) follow-ups and responded to the survey at each of these times will be included in the

analysis. Subsequently, there are fewer students per high school included in the analysis, which

also impacts model specification and is discussed in greater detail in the following sections.

Methodology

In order to understand the contextual effects of school characteristics on the occupational

aspirations of students over time, hierarchical linear modeling (HLM) (Raudenbush & Bryk,

2002) is utilized, which allows regression analysis at multiple levels and specifically the analysis

of students nested in schools. HLM improves the estimates of regression models for individual

66

students and schools in that it borrows strength from similar estimates existing with other

students at other schools. In addition, it allows for the partitioning of the variance and

covariance components among the levels, allowing an understanding of any contextual effects

that may exist by measuring within and between school effects. This analysis includes a 3-level

linear growth model in which level 1 is comprised of individual student growth trajectories for

locus of control, achievement, and occupational aspirations based on observations at 8th, 10th, and

12th grades. These growth parameters are the outcome variables for the level 2 model. Level 2

captures the variation in growth parameters among students within schools, which takes into

account specified student level characteristics. The student level variables at level 2 are the

outcome variables for level 3. Level 3 captures that variation among schools and includes

specified school level characteristics. Data were imported from SPSS 19.0 statistics software

into HLM 6.08 software (Raudenbush, Bryk, Cheong, Congdon, & du Toit, 2004) and models

were estimated at full maximum likelihood (ML) since the number of schools in the model at

level 3 is large and based on the size, ML allows the estimates to converge on the true parameter

values with reduced bias (Raudenbush & Bryk, 2002, p. 408) while also allowing to test for both

alternative variance-covariance structures and alternative specifications of the fixed coefficients

(Raudenbush et al., 2004, p. 71). While the study utilizes nested regression, it also specifies

mediation and as such incorporates the analysis of the potential mediation effects of student

locus of control. All variables and their proposed treatment, multilevel mediation tests and HLM

models are outlined below.

67

Treatment of Data

Missing Values

All missing values in the dataset were specified as missing and showed as a “.” rather

than as a numeric value within SPSS. Since missing data cannot be analyzed within HLM, a

determination on how to treat missing data was made. Initially, listwise deletion was used to

eliminate cases of students who were not in the appropriate grade when the data for the wave

were being conducted. For example, if a student that was part of the dataset in the 8th grade base

year did not progress to the 10th grade two years later but remained in a lower grade, that student

would still be asked to participate in the 1st follow-up, even though they would not be in the 10th

grade. Since this study analyzes the growth of occupational aspirations per year, inclusion of

students who were not progressing through the grades at an annual pace would not allow for an

accurate generalization of annual growth. Thus, such cases were eliminated. Similarly, if cases

were missing school ID or panel weight information, they were also eliminated using listwise

deletion since HLM cannot handle missing data and both the school ID and panel weight

information are necessary for analysis in HLM at level 2 and level 3. Finally, with over 12,000

cases across three waves, there was bound to be missing data among the core variables of

interest. Since HLM does not analyze cases with missing data it was important to understand the

magnitude and underlying reasons for missing data within the dataset. After an initial analysis, it

was determined that over 80% of the cases were missing data, and listwise deletion of all cases

with missing data would reduce the number of cases available to less than 2,000 resulting in a

significant loss of power and potential for invalid inferences since the deleted data may not be

missing completely at random and thus might bias the results and generalizability (Little &

68

Rubin, 1987). After considering how to approach the missing data, multiple imputation was

chosen.

Multiple Imputation

Multiple imputation generates multiple values for each missing value from the student’s

other observed values and in relation to observed data from other students using regression.

While single imputation would fill in missing values, it would underestimate the variability in

the data, causing biased variance estimates and invalid results (Little & Rubin, 1987). There are

assumptions regarding the missing data when using multiple imputation, the primary assumption

being that the data are missing at random (MAR) and that there isn’t any inherent information

about the probabilities of missingness within the missing data (Little & Rubin, 1987; Rubin,

1976). SPSS software produced a set of five imputations using the core variables of interest.

Then, for use within HLM, multiple MDM files were created to reflect the multiple imputation

data. Within HLM, Rubin’s (1987) rules for combining data are followed and the results for

models reported for the combined dataset.

Weighting

Weighting of the dataset was performed in HLM using the weighted variable produced

by NELS for the purpose of analyzing NELS data. In this analysis, the variable (F4PNLWT)

was chosen because it applied to sample members who completed questionnaires in all five

waves of the NELS:88 survey. In NELS, weighting was calculated for the following reasons:

The general purpose of weighting survey data is to compensate for unequal probabilities of selection and to adjust for the effects of nonresponse. Weights are often calculated in two main steps. In the first step, unadjusted weights are calculated as the inverse of the probabilities of selection, taking into account all stages of the sample selection process.

69

In the second step, these initial weights are adjusted to compensate for nonresponse; such nonresponse adjustments are typically carried out separately within multiple weighting cells. This is the process that was applied to weighting NELS:88 data in all rounds. (Ingels, et al., 1994, p. 43)

Variables

General Treatment

When necessary, variables taken from NELS were transformed into new variables in

SPSS. For example, both the father’s and the mother’s occupations were translated into an

occupational prestige score and then these scores were averaged to create the parental occupation

variable used in this analysis. In addition, proxy variables were used when the dataset did not

include a specific measure of interest. For example, in the case of the variable measuring school

level SES, the dataset variable measuring the percentage of students receiving free and reduced

priced school lunches was used as a proxy. Variables may also have been recoded in order for

the data to be meaningful in the analysis. For example, with student level variables such as sex

and minority, the variables were recoded into dummy variables (0,1). Both dependent and

independent variables are outlined below, noting any specific treatment as well as their

relationship to prior research.

Dependent Variables

The theory guiding this study is that schools, by their very nature of being institutions in

which students spend much of their developmental time, have an effect on the development of

occupational aspirations, but that this effect is indirect and mediated by the effect on a students’

locus of control which in turn effects students’ occupational aspirations. In addition, school

context may impact student locus of control which in turn may impact student achievement

70

which may also impact the development of student occupational aspirations. Based on the

theory, prior literature, and hypotheses previously noted, three separate dependent variables are

examined in this study: student locus of control, student academic achievement, and student

occupational aspirations.

Locus of control. A composite variable of 13 separate items was created within

NELS:88 under the label of locus of control. However, extensive review of this composite using

factor analysis has shown that the items which have been grouped together into one variable

actually load onto two distinctly different constructs: self-concept and locus of control.

Furthermore, the reliability of the composite variable was tested and found that the reliability

was increased when the items were separated into two different constructs (self-concept and

locus of control) based on their loadings. In addition, the predictive power of the composite was

analyzed based on its correlation with math, science and reading scores also reported in the

NELS:88 database, which revealed that when the items were separated into two different

constructs, their predictive power was higher (Freidlin & Salvucci, 1995). Based on this

information, a new scaled composite variable has been created for the purposes of this analysis,

which includes only the 5 out of the 13 items which were significant and loaded onto the locus of

control factor noted by Friedlin and Salvucci and are meant to measure student perception of

how chance versus one's own actions affects the way that life unfolds, and not student self-

concept. Table 1 outlines the items included in the locus of control variable used in this analysis.

In relation to the literature on locus of control, this composite variable provides a

measure of a student’s external locus of control, or the extent to which the student believes that

external forces beyond their influence have control over their lives. Responses were measured

71

Table 1 Composite Variable Items for Student Locus of Control

Item Variable Label (BY = Base Year, F1 = 1st follow up, F2 = 2nd follow up)

Item Description

Connections to Research Literature

BYS44C, F1S62C, F2S66C Good luck is more important than hard work Rotter, 1966; Seifert, 2004

BYS44F, F1S62F, F2S66F Every time I get ahead something stops me

Rotter, 1966; Phares,1976; Mortimer, et al., 2002

BYS44G, F1S62G, F2S66G Plans hardly ever work out Rotter,1966; Seifert, 2004

BYS44B, F1S62B, F2S66B I don’t have enough control over my life

Rotter,1966; Ames,1992; Mortimer, et al., 2002;

Mirowsky & Ross, 2003; Seifert, 2004; Schieman and

Plickert, 2008

BYS44M, F1S62M, F2S66M Chance and luck are very important in my life Rotter, 1966; Seifert, 2004

using a Likert scale from 1 to 4, with 1 being “strongly agree” and 4 being “strong disagree.”

With this in mind, lower scores on this measure indicate an external locus of control in which a

student believes they have little control over the events in life. Higher scores indicate an internal

locus of control and the student belief that events in their life are within their control to

influence. This repeated measure was utilized to analyze the change in locus of control from 8th

grade (base year) to subsequent observations in 10th (1st follow up) and 12th grades (2nd follow

up).

Student occupational aspirations. The student occupational aspirations variable

consists of a single item within the NELS:88 which asked students at each wave in the

longitudinal study what kind of work the student expected to do at the age of 30. Students were

given occupational categories from which to choose. These categories were coded numerically

72

from 1-14, but the codes held no inherent value in connection to the occupational categories. In

order to test the hypotheses in this study, the NELS:88 measure of student occupational

aspiration has been transformed into a new variable that reflects a level of occupational prestige

associated with the occupational category by using the Nakao-Treas occupational prestige scale

(1994). As noted in the review of the literature, the Nakao-Treas index it is one of the most

widely used occupational prestige scales because it is one of the most current major national

surveys of occupational prestige and is based on contemporary prestige ratings of the

occupational titles that appeared in the 1980 census which was then updated to reflect the 1990

census information (Hauser & Warren, 1997; Nakao & Treas, 1992, 1994). This index was also

chosen because the scale reflects the timeframe of the NELS:88 dataset utilized in this analysis.

To be clear, occupational prestige scores serve as proxy for occupational aspirations in this

study. For all subsequent analysis and discussion of occupational aspirations in this study, the

basis of that occupational aspiration is no longer a category of occupation, but a corresponding

score on the occupational prestige scale developed by Nakao and Treas.

The use of occupational prestige was chosen as a way to quantify occupational

aspirations based on the literature on status attainment, which addresses both the individual and

societal components of attainment. At the individual level, researchers note that students often

aspire to occupations which will enhance their status (Csikszentmihalyi & Schneider, 2000;

Goyette, 2008; Inoue, 1999), as occupation in the US is a strong determinant of wealth, lifestyle,

and individual status within a community (M. K. Johnson & Mortimer, 2002). At the societal

level, technology requires advanced knowledge on the part of the majority of the population

rather than an elite few. As such, greater mobility across occupational attainment structures is

imperative in order for larger numbers of the population to contribute their knowledge and skills

73

in the technological age (Blau & Duncan, 1967). To sustain the social structure, society needs

individuals to aspire to greater occupational status and subsequent social mobility (Blau &

Duncan, 1967; Haller & Portes, 1973; Inoue, 1999; National Commission on Excellence in

Education, 1983; Sewell et al., 1970; Weiss, 2003).

The occupational prestige scale by Nakao and Treas (1992, 1994) was an interval scale

developed by using the occupational categories found in the 1980 US census and subsequently

updated using the 1990 census occupational categories. Specifically, the Nakao-Treas prestige

scale is based on a nationally representative sample of adults (N = 1,166) who were used to

evaluate the prestige of 740 occupational titles by sorting small cards onto a 9-rung ladder of

social standing with “1” being the lowest social standing and “9” being the highest. Subsamples

were employed to cover all of the occupational categories, with each respondent randomly

assigned to one of 12 subsamples, where they rated 110 titles. The work by Nakao and Treas

follows previous work on prestige. Specifically, the Nakao-Treas process for calculating

prestige scores for occupational titles replicated that used for developing prestige scales by the

National Opinion Research Center (Hodge, Seigel, & Rossi, 1964) in the 1960s. With the ratings

in hand, the authors followed a previously devised formula to convert the ratings over the nine

rungs of the social standing ladder into “12.5 point intervals so that the prestige scores would

have a logical range from 0 (lowest) to 100 (highest)” (Nakao & Treas, 1994, p. 8). Based on

scored occupational titles, scores were then assigned to occupational categories.

Examples of highly rated occupational titles include physicians (86 on the 100 point

scale), physicists (73), and college professors (74). Lower rated occupational titles include

unskilled factory workers (23), delivery drivers (24), and restaurant servers (28). Examples of

mid-level prestige titles include auto mechanics (40), lobbyist (46), kindergarten teachers (55),

74

and chiropractors (57). The mean prestige score is 47.5, roughly the prestige of occupational

titles such as postal worker, insurance adjuster, or accountant. While the Nakao-Treas is a scaled

score from 0-100, student choices within the NELS were limited and do not reflect the entirety of

the Nakao-Treas scale (see Table 2).

Table 2

Variable for Occupational Aspirations in 8th Grade

NELS Category

Choices & Description (For BY 8th grade)

NELS

Category

Nakao &

Treas Prestige

Score Craftsperson 1 39 Farmer/Farm Manager 2 44 Housewife 3 51 Laborer/Farm Worker 4 28 Military/Police/Security 5 56 Business/Managerial/Professional 6 57 Business Owner 7 52 Technical Person 8 59 Sales/Clerical 9 38 Science/Engineering 10 76 Service Worker 11 35 Other 12 50 Not Working 13 0 Don’t Know 14 NA

In fact, the categorical choices available to students differ from BY in 8th grade to the 1st

follow up in 10th grade, with fewer categories from which to choose, and a lower occupational

prestige ceiling, given the choices. In the 1st follow up, students were given a slightly broader

range of categories from which to choose, increasing their occupational prestige possibilities,

with the highest category available being that of a professional such as a doctor, with a prestige

75

score of 86 as opposed to the highest score on the BY which is a 76 for a scientist or engineer

(see Table 3).

Table 3

Variable Changes for Occupational Aspirations in 10th and 12th Grades

NELS Category Choices & Description

(for 1st follow up 10th grade and 2nd follow up 12th grade)

NELS

Category

Nakao & Treas

Prestige Score

Office Worker 1 42 Tradesperson 2 36 Farmer 3 40 Full Time Homemaker 4 51 Laborer 5 28 Manager 6 57 Military 7 56 Machine Operator 8 34 Professional 1 (architect, engineer) 9 71 Professional 2 (doctor) 10 86 Proprietor/Owner 11 52 Protective Services 12 55 Sales 13 38 School Teacher 14 65 Service 15 35 Technical 16 59 Never Worked 17 0

While this may not be a perfect measure in terms of broad occupational choices, the

transformation of the variable from categorical to occupational prestige allows for increased

knowledge not only of an occupational category to which students aspire, but an understanding

of the level of prestige and social standing associated with those aspirations. Since this measure

was repeated in each wave of the NELS:88, student responses from the 8th grade base year

76

through the 10th and 12th grades will be utilized to analyze the change in occupational aspirations

over time. However, due to the shift in categorical choices in the NELS as noted above, using

two different scales may mean that any change in aspirations found within the study may be

somewhat inflated in that students have a wider range of choices in the subsequent follow ups to

the base year. To eliminate inflated change in aspirations over time due to differences in scales,

this variable has been created by reducing the 10th and 12th grade categories into their 8th grade

equivalents and measured accordingly. Table 4 shows the category changes, the largest being

the reduction from Professional 2 (doctor) in the 10th and 12th grade scales, which has a prestige

score of 86, to the 8th grade equivalent category of Science/ Engineering which has a prestige

score of 76, a difference of 10 points on the prestige scale. Overall, there are five categories

which experience small increases in prestige, for a total increase of 14 prestige points and three

that experience decreases for a total decrease of 22 prestige points, leaving a net loss of 8 points.

An additional limitation is that the categories tend to have prestige scores that cluster around the

mean (50), and thus may not be as representative of the overall aspirational prestige of students

as they might otherwise have been with the inclusion of additional categories in the NELS

questionnaire. Table 5 provides an overview of the occupational aspirations variable created.

Academic achievement. Research shows that strong academic achievement enhances

future occupational opportunities (Inoue, 1999) and provides youth with strong advantages in

occupational attainment and earnings (Parcel & Dufur, 2001; Rosenbaum, 2001) while weak

academic achievement and academic attainment predicts lower earnings (S. R. Miller &

Rosenbaum, 1997). Still, students are often unaware of these relationships, and may not factor

their academic performance (specifically if weak) into their occupational aspirations

77

Table 4 Prestige Score Differences in BY and F1/F2 Scale for Student Occupational Aspirations

NELS Category

Description For F1 and F2

NELS

Category

Nakao & Treas Prestige

Score (F1/F2)

NELS Category

Description For BY 8th grade

equivalent

Nakao & Treas

Prestige Score (BY)

Score Difference

from F1/F2 to

BY Office Worker 1 42 Clerical 38 -4 Tradesperson 2 36 Craftsperson 39 +3 Farmer 3 40 Farmer/Farm Manager 44 +4 Full Time Homemaker 4 51 Housewife 51 0 Laborer 5 28 Laborer/Farm Worker 28 0

Manager 6 57 Business/Managerial/ Professional 57 0

Military 7 56 Military/Police/Security 56 0 Machine Operator 8 34 Service Worker 35 +1 Professional 1 (architect, engineer) 9 71 Science/Engineering 76 +5 Professional 2 (doctor) 10 86 Science/Engineering 76 -10 Proprietor/Owner 11 52 Business Owner 52 0 Protective Services 12 55 Military/Police/Security 56 +1 Sales 13 38 Sales/Clerical 38 0

School Teacher 14 65 Business/Managerial/ Professional 57 -8

Service 15 35 Service Worker 35 0 Technical 16 59 Technical Person 59 0 Never Worked 17 0 Not Working 0 0 (-8) Total

78

Table 5 Variable for Student Occupational Aspirations

Variable Label

(noted from Base Year,1st follow up, 2nd

follow up)

New Variable

Label (for use in this

analysis)

Description

Type of Variable

Treatment

BYS52, F1S53B, F2S64B

OCC30 Occupation

student aspire to have at age 30

Categorical

Transformed reported occupational categories

into occupational prestige scores

(Rosenbaum, 2001), thus having misaligned aspirations that may require higher levels of

education of which they may be unaware and for which they may be unprepared, especially if

they have weak academic performance. Given the relationship between achievement and

aspirations noted above and in the previous review of the literature, it is hypothesized that

schools might impact student occupational aspirations through the impact of school contextual

characteristics on student achievement via their impact on student locus of control.

For this analysis, math scores taken from the NELS:88 serve as the proxy for student

academic achievement. As part of the NELS:88 survey, students were given a battery of

multiple choice tests in various areas, including mathematics. In the base year, all students

received the same test. Then, multiple forms of the test were developed for follow-up testing in

subsequent waves based on track and coursework exposure to account for any floor or ceiling

effects (more correct answers to questions that were too easy for advanced math students or more

guessed answers to difficult test questions by less advanced math students). The appropriate

follow-up tests were given at 10th grade and again in 12th grade based on students’ academic

79

track and exposure to previous mathematics coursework. Scores were then scaled using item

response theory (IRT) scoring in order to compare scores from tests with varying degrees of

difficulty. According to the NELS:88,

Item Response Theory (lRT) was employed to calculate scores that could be compared regardless of which test form a student took. A core of items shared among the different test forms made it possible to establish a common scale. IRT uses the pattern of right, wrong, and omitted responses to the items actually administered in a test form, and the difficulty, discriminating ability, and “guess-ability” of each item, to place each student on a continuous ability scale. It is then possible to estimate the score the student would have achieved for any arbitrary subset of test items calibrated on this scale. (Ingels et al., 1994, p. H-32)

In order to analyze growth in student academic achievement, the theta score variables (Rock,

2012) from 8th, 10th, and 12th grade have been utilized in this study, which is noted in Table 6.

Table 6 Variable for Student Academic Achievement

Variable Label

(noted from Base Year,1st follow

up, 2nd follow up)

New Variable


analysis)

Description

Type of Variable

Treatment

BY2XMTH, F12XMTH, F22XMTH

ACHIEVE

Student math scores (proxy for

academic achievement)

Scaled None

Independent Variables

Student (Level 2) variables were specifically chosen based on the cited literature which

outlines their significance in relation to student occupational aspirations and locus of control.

They have been included in the model either as predictor variables, as is the case with academic

80

track in relation to locus of control, or in order to control for their effects, as is the case with the

student characteristic variables such as socioeconomic status (SES), race, gender, and parental

occupation.

Curriculum track. The literature on the effects of academic tracking on students is rich,

and as noted in the review of the literature, types of tracking can be persistent (Alexander &

Entwisle, 2000; Entwisle & Alexander, 1993; M. K. Johnson & Mortimer, 2002; Paul, 1997a,

1997b), may be barriers to future opportunity (Hallinan, 1996, 2000; Orfield, 1997) and may

provide students with a perception of self that may also play a motivational role in their

aspirational development (Alwin, 1994; Lent et al., 1996; O'Rand, 2000).

For inclusion in this study, two types of variables were considered: the student reported

variable of their curriculum track (available at the 1st and 2nd follow ups) and the school reported

variable of the student’s curriculum track (available at the 2nd follow up). From these two, the

school reported variable was eliminated from consideration for two reasons: first, while the

school reported track may be a more accurate assessment of the student’s academic track, using

the school reported variable is not a reflection of the student’s knowledge of their curricular

track, which is the reason for including academic track in the analysis. Using the school reported

track will not provide an adequate understanding of whether the student is aware of the track

they are in and then whether their track influences their locus of control or academic

achievement. Second, the school reported variable was missing more information than the

student reported variable. Instead, the student reported track from the second follow up was

included (F2S12A), which asked students to describe their high school program, which provides

some insight as to whether the student was aware of their curricular track and subsequently how

that track might influence their locus of control and academic achievement.

81

The academic track variable was originally coded as five categories where 1 and 2

represented academically oriented programs and 3 through 15 included vocationally related

programs. Since this analysis was interested in whether vocational coursework, as indicated by a

vocational track, might influence student locus of control, the variable was recoded as a dummy

variable as follows: 0 = academically oriented track; 1 = vocationally related track.

SES. Socioeconomic status (SES) often plays a role in what many students consider

realistic aspirations. In addition to the previous discussion of status attainment noted in the

review of the literature, Bourdieu’s concept of social reproduction (Bourdieu & Passeron, 1977)

laid the groundwork for further research underscoring the effect of SES on educational

opportunities, occupational knowledge, role models, material and network resources, and

perceived barriers to occupational opportunities (Macleod, 1987; Orfield, 1997), all of which

influence occupational aspirations (Hansen & McIntire, 1989; Schoon, 2001; Schulenberg,

Vondracek, & Crouter, 1984). Student level SES also influences the quality of academic

pursuits and perceptions of ability (Franklin, 1995; Inoue, 1999) which in turn may affect the

occupations to which a student aspires (Parcel & Dufur, 2001). In addition, research on locus of

control has found that student SES often indicates a locus of control orientation, with higher SES

indicating a more internal locus of control (Bartel, 1971; Battle & Rotter, 1963; Stipek, 1980).

Since this study is focused on the impact of school contextual characteristics on student

occupational aspirations via student locus of control, it does not include SES as a predictor

variable. However, given its importance to occupational aspirations, it has been included as a

control variable. To control for student SES in this study, the variable reporting SES from the

2nd follow up of the NELS:88 will be utilized (F2SES1), since this variable was derived from the

82

base year parental data, but is the most accurate, as it corrects for a NELS transposition error

found in earlier waves (CIC, 1994). This is a standardized score.

Race. Similar to SES, research on race and occupational aspirations discusses the

concept of reproduction of race in occupational choices and highlights that occupational

aspirations appear to be related to perceptions of opportunities (Flowers et al., 2003) and the

reality of racism in the marketplace (Cook et al., 1996). Minority students may be more likely to

perceive barriers to occupational choices based on their race resulting in prematurely foreclosing

career options (Fouad, 2007), which impacts occupational aspirations.

Race is included in this analysis as a control variable. For the purposes of this analysis,

the variable for race found in the NELS:88 database (BYS31A) from the base year (8th grade)

was recoded as a dummy variable as follows: 0 = White, 1 = minority.

Gender. Studies differ on whether gender matters to the development of occupational

aspirations of students in high school, with some studies showing that both genders aspire to

professional occupations by the age of 30 (Rasinski, Ingels, Rock, Pollack, & Wu, 1993) while

others show that females aspire to professional occupations in a greater percentage than men

(Schneider & Stevenson, 1999), men aspire to higher status occupations (Inoue, 1999), or that

male and female high school students aspire to different, more gender specific occupations

(Armstrong & Crombie, 2000; L. Miller & Bud, 1999; Schneider & Stevenson, 1999; Schoon,

2001). Inclusion of this variable allows controlling for it.

In this analysis, the control variable for student gender was taken from the NELS:88

database (BYS12) from the base year (8th grade) and was coded as a dummy variable, with 0 =

male and 1 = female.

83

Parents occupation. Adult models are an essential ingredient in developing

occupational aspirations (Csikszentmihalyi & Schneider, 2000). Research on the effect of

parental occupations on the occupational aspirations of students has a long history (Blau &

Duncan, 1967; Haller & Portes, 1973; Haller & Woefel, 1972; Sewell et al., 1970) with more

recent research positing that parental occupations have a strong influence on occupational and

educational aspirations and attainment (Kalmijn, 1994; Lampard, 1995; Updegraff, 1996).

Orfield’s work (1997) showed significant statistical correlations between parents with less

education and lower status jobs and students who chose vocational programs in high school

rather than higher academic programs. Such student choices are an indication of occupational

aspirations.

For the purposes of this analysis, both the occupational status of student’s mothers and

fathers taken at the base year (8th grade) were recoded to the Nakao and Treas occupational

prestige scale (1994) and the mean of the combined occupational status of the student’s parents

utilized (see Table 7). While the Nakao and Treas Prestige scale ranges from 0-100, the

categories offered in the NELS do not reflect the entirety of the prestige scale.

School (Level 3) variables were chosen for the same reasons as student level variables:

based on the literature citing the effect of these level 3 variables on locus of control and on

occupational aspirations.

School SES. The potential contextual effects of school SES on student locus of control

and occupational aspirations is couched in the research discussed in the literature review.

Hanson (1994) noted that student inequalities such as lower SES reinscribed in educational

structures (schools with lower SES) may inhibit development of higher occupational aspirations.

84

Table 7 Variable for Parental Occupation

Variable Label

(noted from Base Year)

New Variable


analysis)

NELS Category

Description

NELS

Category

Nakao &

Treas Prestige

Score BYS4OCC, BYS7OCC PAROCCUP Clerical 1 32 Craftsperson 2 39 Farmer/Farm Manager 3 44 Homemaker 4 51 Laborer 5 28 Manager/Administrator 6 57 Military 7 56 Machine Operator 8 34 Professional 1

(architect, engineer) 9 71 Professional 2 (doctor) 10 86 Proprietor/Owner 11 52 Protective Services 12 55 Sales 13 38 School Teacher 14 65 Service 15 35 Technical 16 59 Never Worked 17 0

Similarly, schools with higher SES also have higher levels of student achievement, which is in

turn linked to higher student occupational aspirations and locus of control. Research points to

the importance of the school’s role in providing information on career options and opportunities

to students (Kelly & Lee, 2001; McWhirter et al., 2000; Mortimer et al., 2002; Orfield, 1997),

but school SES may impact the school’s ability to provide career and guidance resources to

students. School SES may also impact the school’s ability to hire quality teachers, which might

also impact student locus of control (Plucker, 1998) and occupational aspirations. If not all

85

schools have the same resources to assist students (Flowers et al., 2003; Gamoran, 1996), then

the level of available resources and the potential subsequent effect on student locus of control

and occupational aspirations may be related to school SES.

Given this, the level of school SES was included. A proxy variable for the SES of the

high school was created by determining the percentage of students receiving free or reduced

priced lunch (FRPL) as found in NELS:88 in the 1st follow up (F1C30A). Tenth grade was

chosen since it is the midpoint through a four-year high school program and would be most

representative of the school SES for the bulk of time over which any changes in the dependent

variables would occur. While schools report the percentages of students receiving FRPL, NELS

recoded these percentages into categories, which, for the purposes of this study were transformed

into a dummy variable. The treatment for this variable is noted in Table 8.

School climate. As noted by Rumberger (1995), school organization and climate are

important for fostering achievement and internal locus of control, which, as hypothesized in this

study, may in turn affect student occupational aspirations. Further review of the literature

acknowledges that school climate may be reflected by a variety of factors. Given that, variables

related to teachers and the organization of their classrooms are of specific interest and will be

used to define school climate in this analysis, as teachers are the most consistent interface with

students on a daily basis. Specifically, previous research shows that the interest and commitment

of teachers to the students in their classrooms and schools (V. E. Lee & Bryk, 1989), students

receiving encouragement from teachers within their schools (Patrick et al., 2007; A. M. Ryan &

Patrick, 2001) and even the organization of classrooms or how lessons are taught (Franklin,

1995; A. M. Ryan & Patrick, 2001), are important factors that can affect student perceptions of

contextual supports which in turn affects student self-perceptions such as locus of control

86

Table 8 Variable for School Level SES

Variable Label

(noted from 1st follow up)

New Variable Label

(for use in this analysis)

Description

Type of Variable

Treatment

F1C30A SCHOOLSES

Measures the SES of the school

through a proxy variable – the

percentage of the student

population receiving Free and Reduced

Priced Lunches (FRPL)

Dummy

The percentage of students receiving FRPL was recoded by NELS

into a categorical variable: 0 = 0%

1 = 1-10% 2 = 11-50% 3 = 51-100%

Which was transformed into a dummy variable for

use in this analysis: 0 = 50% or less receive

FRPL 1 = 51% or more receive

FRPL (Daniels & Araposthasis, 2005; Plucker, 1998) and thus, as hypothesized, occupational

aspirations. Inclusion of a school climate variable is one way to measure whether schools have

an effect on student locus of control and, indirectly, occupational aspirations.

For inclusion in this analysis, a composite variable within NELS:88 was considered.

This composite is made up of 13 separate items within the NELS:88 to help define the climate of

a school as answered by a school counselor or someone serving the role of a counselor within a

school. However, a detailed review of this composite (Freidlin & Salvucci, 1995) has shown that

the items which have been grouped together under the NELS:88 label of “school climate” did not

all load onto the construct of school climate. Those items that did load onto the school climate

variable, once tested, had high internal consistency (Cronbach’s alpha of .868), but as a

87

composite reduced the predictive power to such a degree as to make the composite “useless”

(Freidlin & Salvucci, 1995, p. 41). Table 9 lists each of the items noted in the NELS:88 school

climate composite and their subsequent loadings onto the school climate construct.

Due to the reduced power of these items as a composite, Freidlin and Salvucci (1995)

recommend using these items individually within an analysis. While Freidlin and Salvucci tested

this composite using only the base year (8th grade) variables, the school climate composite

created within NELS:88 remains relatively unchanged for the 1st follow up (10th grade), with

differences in the composite being the omission of four items (BYSC47J through BYSC47M)

that were measured during the base year but were not measured in the 1st follow up, as recorded

in the table. Since all other items were the same, no significant changes in factor loadings onto

the construct of school climate were anticipated. Therefore, in keeping with their findings and

the literature on school climate, this analysis combined the following individual items from the

10th grade 1st follow up, (also noted in Table 9), into one composite variable to measure school

climate: teacher morale and classroom structure. Teacher attentiveness was not included since it

was not measured again in the 1st follow up. Tenth grade was chosen since it is the midpoint

through a four-year high school program and would be most representative of the school climate

for the bulk of time over which any changes in the dependent variables would occur. Since both

of these items are based on a 5 point Likert scale, they were kept as scaled scores for this

analysis.

Academic press. Whether teachers press students to achieve is the topic of much

research. As noted more specifically in the review of the literature for this analysis, the quality

of instruction and the interest and commitment of teachers is important not only for encouraging

student achievement, but for encouraging internal locus of control (Plucker, 1998; Stipek, 1980).

88

Table 9 School Climate Composite Variable Items

Item Variable Label (noted from Base Year and 1st follow up)

Item Description

Factor Loading (from Base Year-Friedlin & Salvucci 1995)

Included in current analysis?/ Type of Scale

Connections to Research Literature

BYSC47A, F1C93M

There is conflict between teachers

and administrators .329 No

BYSC47B, F1C93A

Discipline is emphasized at the

school

.601 No

BYSC47C, F1C93B

Students place high priority on

learning .448 No

Eccles & Wigfield, 1985; Plucker, 1998; Yazzie-Mintz, 2009

BYSC47D, F1C93C

Classroom activities are

highly structured .640

Yes – School Climate

(5 pt Likert

scale)

Barr & Dreeben, 1983; Bradley &

Gaa, 1977; Gamoran &

Berends, 1987; Gamoran & Mare, 1989; Kerckhoff, 1986; A. M. Ryan & Patrick, 2001;

Stipek, 1980

BYSC47E, F1C93D

Teachers press students to achieve .811

Yes – Academic

Press

(4 point Likert scale)

Eccles & Wigfield, 1985; Flowers, et al., 2003; Plucker,

1998; Seifert & O'Keefe, 2001

BYSC47F, F1C93E

Students are expected to do

homework .767

Yes – Academic

Press (4 point

Likert scale)

Flowers, et al., 2003; Plucker,

1998

(Table continues)

89

BYSC47G, F1C93F

Teacher morale is high .621

Yes – School Climate

(5 point

Likert scale)

Daniels & Araposthasis,

2005; Eccles & Wigfield, 1985;

Klem & Connell, 2004; Plucker,

1998 BYSC47H,

F1C93K

Teachers have a negative attitude toward students

.282 No

BYSC47I, F1C93L

Teachers have difficulty

motivating students

.156 No

BYSC47J,

The school day is structured .640 No

BYSC47K

Rule deviation is not tolerated .560 No

BYSC47L

Environment is flexible .257 No

BYSC47M

Teacher responds to individual needs .598

No (would be included but

does not have corresponding variable in the 1st follow up

– F1)

Flowers, et al., 2003; Franklin, 1995; Patrick, et

al., 2007; R. Ryan & Deci, 2000;

Seifert & O'Keefe, 2001

BYSC47N, F1C93H

School emphasizes sports .165 No

BYSC47O, F1C93I

Students compete for grades .244 No

90

If achievement and perceptions of control are related (Bartel, 1971; J. S. Coleman et al., 1966;

Graham, 1991; Mau & Bikos, 2000; Murasko, 2007; Neild, 2009; Stipek, 1980) and reinforce

each other (Bradley & Gaa, 1977; Ross & Broh, 2000; Stipek, 1980), then understanding how

schools encourage achievement (i.e., through teachers) is important to understanding how they

may indirectly influence student occupational aspirations via their influence on student locus of

control.

For the purposes of this study, school level variables within NELS:88 associated with

academic press were considered. Initially included as part of a composite variable measuring

school climate, these items were included based on the recommendations of Freidlin and

Salvucci (1995) specified in the previous variable on school climate. The three specific items

measuring academic press noted in Table 9 include: teachers pressing students to achieve,

students expected to do homework and students placing a high priority on learning. While all

three items address academic press, only two items (teacher pressing students to achieve and

students expected to do homework) focus on what schools do to encourage achievement while

the other item (students placing a high priority on learning) addresses how students may respond

to that encouragement. Thus, only the two school-oriented items from the 1st follow up (10th

grade) were combined into a composite variable and included as measures of academic press in

this analysis. Tenth grade was chosen since it is the midpoint through a four-year high school

program and would be most representative of the academic press for the bulk of time over which

any changes in the dependent variables would occur. Since both of the items are based on a 4

point Likert scale (forced response), the composite was kept as a scaled score for this analysis.

Counseling support. With a burgeoning number of students with misaligned

aspirations, the role of counselors is emerging as an important one (Csikszentmihalyi &

91

Schneider, 2000; Rosenbaum, 2001; Schneider & Stevenson, 1999). Although counselors have

been reticent in counseling students toward non-college pursuits (Rosenbaum, Miller & Krie,

1996; Rosenbaum et al., 1997), and students have not often seen counseling as particularly

helpful, research confirms that access to fully implemented guidance programs have positive

relationships with achievement and occupational development (Borders & Drury, 1992; Cicourel

& Kitsuse, 1963; Gerler, 1985; Kerckhoff, 2000a; Whiston & Sexton, 1998). As such, including

a measure of whether schools have counseling support available to students is important to

understanding how such access may indirectly influence student occupational aspirations via

their influence on student locus of control.

For inclusion in this study, variables within NELS:88 1st follow up (10th grade) indicating

counseling support were considered. While NELS:88 provides a measure (F1C41J) of the

number of guidance counseling faculty within a school, which would indicate whether the school

provided access to formal counseling support to students, the variable is categorical and

respondents were not given the option of answering “none” if their school did not have any

guidance counselors. Instead, the category available was “0-5,” thus negating the ability to

distinguish between those schools with counselors (1 or more) and those without counselors (0).

There are no other variables within the 1st follow up that clearly indicate whether the school

provided student access to counseling support. Since students attended the same high school in

the 2nd follow up (12th grade), variables within the 2nd follow up (12th grade) indicating school

provision of a formal counseling program for students were considered. Unlike the 1st follow up,

the 2nd follow up includes a specific item (F2C35K) that asks whether the school has a formal

guidance counseling program as well as an item (F2C36K1) that asks the number of full-time

guidance counselors employed by the school and thus indicating the availability of formal

92

counseling support available for students. The second item differs from its 1st follow up

counterpart in that it offers a category choice of “none” for schools without full time counselors.

While this variable offers insight into whether a counseling program has full time dedicated staff,

it does not eliminate the possibility of a formal counseling department available to students

employing multiple part time counselors rather than full time counselors. Thus, the best way to

measure counseling support is via the 2nd follow up (12th grade) variable (F2C35K) which asks

whether the school has a formal guidance counseling department. Since the variable was already

coded categorically with 1 = “yes” and 2 = “no” to indicate whether there was a formal guidance

counseling department at a student’s high school, for the purposes of clear analysis the categories

were just recoded as 0 = “yes” and 1 = “no” (see Table 10).

Table 10 Variable for School Level Counseling Department

Variable Label

(noted from 2nd follow up)

New Variable Label


Description

Type of Variable

Treatment

F2C35K NOCOUNSEL

Indicates whether there is a formal

guidance counseling

department at a student’s high

school

Dummy

Recoded categories in the original variable to the

following for more clear analysis: 0 = Yes 1 = No

Year. In order to analyze growth over time in the three dependent variables student

occupational aspirations, student locus of control, and student academic achievement, a “Year”

93

variable was created that would allow for the analysis of these variables in each of the three

waves for each student (see Table 11). This variable was coded for each wave, with “0” = initial

wave measure, which was taken in 8th grade, “2” = 2nd wave measure which was taken two years

later in 10th grade, and “4” = 3rd wave measure which was taken four years after the initial wave

in 12th grade. The data were then restructured so that each student had a measure for each year.

Table 12 provides a sample representation of the restructured data in which there are three data

entries for each student ID, since the student has three measures for each of the dependent

variables. Coding the year variable in this manner allows for interpretation of the results in

relation to annual growth. All the variables within the study are noted in Table 13

Table 11 Variable for Analyzing Growth Over Time

Variable

Label

New Variable Label


Description

Type of Variable

Treatment

----- YEAR

Indicates in which wave the student

measure was taken

Categorical

Restructured student data to show three lines per

student in order to capture the multi-wave data for the dependent variables.

94

Table 12

Sample of Restructured Data Showing Data for Each Wave in the Analysis

ID SCH ID YEAR ACHIEVE OCC30 LOC

7898401 81 0 60 4 38.76 7898401 81 2 42 3 48.12 7898401 81 4 42 3 47.14 7898402 81 0 52 3 34.60 7898402 81 2 36 3 37.64 7898402 81 4 64 2 31.53 7898406 81 0 50 3 47.50 7898406 81 2 38 3 58.94 7898406 81 4 40 3 63.61 7898407 81 0 51 3 43.13 7898407 81 2 71 3 49.54 7898407 81 4 86 3 57.71

95

Table 13 Complete Variable Table Variables Description (where R is the individual student) Dependent Variable

(BY = Base Year, F1 = 1st Follow up, F2 = 2nd Follow up)

OCC30 Occupation R aspires to have at the age of 30 (based on the occupational prestige scale by Nakao & Treas, 1994)

LOC ACHIEVE

R’s locus of control. Scaled score reporting how much control R feels s/he has over life and the future. R’s academic achievement over time (proxy measure: math score).

Independent Variables

Level 1 Variable YEAR Level 2 Variables

Repeated measures for R taken at 8th, 10th and 12th grades. Coded as 0,2,4 respectively, so outcomes reflect yearly growth.

SES Socioeconomic status of R (SES from BY)

FEMALE Gender of R (0 = male; 1 = female) from BY

MINORITY Race of R (0 = white; 1= minority) from BY

PAROCCUP VOCTRACK

Occupational status of R’s parents from BY (based on the occupational prestige scale by Nakao & Treas, 1994) R’s HS academic track in F1 (0 = academic; 1 = vocational)

Level 3 Variables LOSCHSES ACADPRESS

SES of HS determined by % of students receiving free or reduced priced lunch (FRPL). Dummy variable (0 = 50% or less FRPL; 1 = 51% or more FRPL) Teachers at R’s high school encourage high academic achievement, Scaled score taken from F1.

NOCOUNSEL There is no formal guidance counseling department at R’s high school taken from F2 (0 = formal guidance counseling dept, 1 = no formal guidance counseling dept)

POSCLIMATE Climate at R’s high school is positive. Scaled score taken from F1.

96

Descriptive Statistics

A standard visual screening and subsequent analysis of the descriptive statistics show that

the variables most generally conform to normality, with means, standard deviations and values

within an acceptable range (see Table 14). While multiple imputation was used to replace

missing values, performing due diligence regarding the acceptable range of values ensured that

the process did not create any outer range values. The one exception to this is the variable

Table 14 Descriptive Statistics

Variable N M SD Min Max Level 1 OCC30 28302 60.20 13.38 0.00 76.00 LOC 28302 3.03 0.50 1.00 4.00 ACHIEVE 28302 51.30 10.08 24.87 80.67 Level 2 FEMALE 9434 0.53 0.50 0.00 1.00 MINORITY 9434 0.31 0.46 0.00 1.00 SES 9434 0.068 0.78 -2.43 2.75 PAROCCUP 9434 46.38 11.40 0.00 86.00 VOCTRACK 9434 0.15 0.36 0.00 1.00 Level 3 LOSCHSES 1315 0.10 0.30 0.00 1.00 CLIMATE 1315 3.88 0.60 2.00 5.00 ACADPRES 1315 3.18 0.65 1.00 4.00 NOCOUNSE 1315 0.12 0.32 0.00 1.00

measuring the occupational aspirations of students at 30 (OCC30). When the data were first

explored in a visual graph using SPSS, most all variables looked normal, with a normal

distribution. However, OCC30 was somewhat negatively skewed and leptokurtic. While it is

97

generally accepted that certain violations of normality can impact the estimation of variable

effects (p-values) and standard errors and increase Type II errors, HLM is generally robust to

violations of normality, especially with larger samples sizes (Kulikowich & Edwards, 2007;

Maas & Hox, 2004), which is the case in this study.

HLM Models

Based on the sampling method discussed in Chapter III and in keeping with the need for

proper specification (Gamoran, 1996), the variables tested in the following models have been

grand mean centered. In theory, group mean centering the variables would allow for an analysis

of the effects of school context between students within a school. However, the given the small

number of students in the study that come from the same school, such an analysis would not be

meaningful in that any inferences would have been drawn from an unbalanced sample. Grand

mean centering the variables allows for an analysis of students across all schools and provides a

significantly larger and more balanced sample from which to draw meaning (Raudenbush &

Bryk, 2002).

The following growth models will be estimated in order to answer the first research

question noted in the previous section. The first model will provide an initial understanding of

change in student locus of control from 8th through 12th grades (model 1). The second model

will provide an initial analysis of change in occupational aspirations from 8th grade through 12th

grade as a baseline understanding of student aspirations (model 2) and the third model (model 3)

will provide an initial analysis of student academic achievement from 8th grade through 12th

grade. The fourth model (model 4) will provide an understanding of whether changes from 8th

grade through 12th grade in student locus of control and student academic achievement influence

98

change in student occupational aspirations. All of the models will enrich the subsequent

mediation models analyzing whether school characteristics impact student occupational

aspirations via student locus of control and academic achievement from 8th grade through 12th

grade.

Model 1: Growth Model for Student Locus of Control

In this model, student locus of control (LOCtij) is estimated, and specifically the how

student locus of control changes over the course of four years, from 8th grade through 12th grade.

Level 1 specifies repeated measures of locus of control for students. In this model, there are no

predictors at level 2 or level 3 in order to analyze the variance in student locus of control within

and between schools.

Level-1 (LOC)tij = π0ij + π 1ij(YEAR)tij + e tij

Level-2 π0ij = β00j + r0ij π1ij = β10j + r1ij Level-3 β00j = γ000 + u00j β 10j = γ100 + u10j Ytij = predicted locus of control of student i in school j at time t π0ij = predicted average locus of control of student i in school j while in the eighth grade π 1ij = predicted per year change in the locus of control of student i in school j over the course of four years in high school

Model 2: Growth Model for Student Occupational Aspirations

In this model, student occupational aspirations (OCC30tij) is estimated, and specifically

the how student occupational aspirations change over the course of four years, from 8th grade

99

through 12th grade. For measuring change over time, the term Year is used. Level 1 specifies

repeated measures of occupational aspirations for students. In this model, there are no predictor

variables in order to analyze the variance of student occupational aspirations within and between

schools.

Level-1 (OCC30)tij = π0ij + π 1ij(YEAR)tij + e tij Level-2 π0ij = β00j + r0ij π1ij = β10j + r1ij Level-3 β00j = γ000 + u00j β10j = γ100 + u10j Ytij = predicted occupational aspirations of student i in school jat time t π0ij = predicted average occupational aspirations of student i in school j while in the eighth grade π 1ij = predicted per year change in the occupational aspirations of student i in school j over the course of four years in high school

Model 3: Growth Model for Student Academic Achievement

In this model, student academic achievement (ACHIEVEtij) is estimated, and specifically

how student academic achievement changes over the course of four years, from 8th grade through

12th grade. For measuring change over time, the term Year is used, allowing the estimation of

initial academic achievement and growth in academic achievement. Once again, there are no

predictor variables in order to understand the initial variance in academic achievement within

and between schools.

Level-1 (ACHIEVE)tij = π0ij + π 1ij(YEAR)tij + e tij Level-2 π0ij = β00j + r0ij π1ij = β10j + r1ij

100

Level-3 β00j = γ000 + u00j β10j = γ100 + u10j Ytij = predicted academic achievement of student i in school j at time t π0ij = predicted average academic achievement of student i in school j while in the eighth grade π 1ij = predicted per year change in the academic achievement of student i in school j over the course of four years in high school

Model 4: Full Growth Model for Student Occupational Aspirations

In this model, student occupational aspirations (OCC30tij) is estimated, and specifically

the how change in student locus of control (LOC tij) and student academic achievement

(ACHIEVEtij) impact change student occupational aspirations change over the course of four

years, from 8th grade through 12th grade. For measuring change over time, the term Year is used.

Level 1 specifies repeated measures of occupational aspirations for students as predicted by

change in locus of control and academic achievement. In this model, there are no student or

school level characteristics in order to understand the unconditional relationship between growth

in student occupational aspirations, locus of control and academic achievement.

Level-1 (OCC30)tij = π0ij + π 1ij(YEAR)tij + π 2ij(ACHIEVE)tij + π 3ij(LOC)tij + e tij Level-2 π0ij = β00j + r0ij π1ij = β10j + r1ij π2ij = β20j π3ij = β30j Level-3 β00j = γ000 + u00j β 10j = γ100 + u10j β 20j = γ200 β 30j = γ300

101

Yti = occupational aspirations of student i in school j at time t γ000 = average occupation aspirations of student i in school j while in the eighth grade γ100 = per year change in the occupational aspirations of student i in school j over the

course of four years in high school γ200 = change in occupational aspirations of student i in school j over the course of

four years in high school associated with a one unit change in academic achievement

γ300 = change in occupational aspirations of student i in school j over the course of four years in high school associated with a one unit change in locus of control

Mediation Testing

While the theoretical model posits that the characteristics of the school environment are

expected to affect student occupational aspirations via student locus of control (as noted in

Figure 4), measuring the extent to which locus of control actually mediates this relationship is

key. Baron and Kenny (1986) noted that “mediators explain how external physical events take

on internal psychological significance” (p. 1176) and are the most cited for tests of mediation for

traditional mediation models in which running three regression models provide evidence for

mediation, if it exists (Iacobucci, 2008). Zhang’s diagram (Zhang, Zyphur, & Preacher, 2009) of

Baron and Kenny’s (1986) work on mediation depicts traditional mediation models and tests to

determine whether mediation is significant.

Baron and Kenny (1986, p. 1176) proposed several conditions that needed to be met for

a variable to function as a mediator: (Path c) in which the independent variables influence the

outcome variable when the mediator is absent; (Path a) in which variations in the levels of the

independent variables significantly account for variations in the mediating variable; (Path b)

variations in the mediator significantly account for variations the outcome variable; and (Path c’)

which tests the influence of the independent variables on the outcome variable but includes the

mediator. According to Baron and Kenny, when paths a and b are controlled, a previously

102

significant relationship between the independent and dependent variables is no longer significant,

with the strongest demonstration of mediation occurring when Path c is zero.

Figure 4. Baron & Kenny (1986) Mediation Models.

While this study employs a three level nested regression model (MacKinnon, 2008;

Pituch, Murphy, & Tate, 2010; Zhang et al., 2009) the same process applies, but instead of

running the classic mediation tests for a single level regression, multilevel mediation tests will be

conducted and analyzed for indirect effects.

Below are general specifications for applying mediation testing to multilevel design to

this study (MacKinnon, 2008) (Figure 5).

1. Level 1: Ytij = π0ij + e tij Level 2: π0ij = β00j + r0ij Level 3: β00j = γ000 + cX 0ij + u00j Figure 5. Path c

103

This test analyzes whether there is a significant direct relationship (c) between the school

characteristics (X) and student occupational aspirations (Y)

(Figure 6).

2. Level 1: Mtij = π0ij + e tij Level 2: π0ij = β00j + r0ij Level 3: β00j = γ000 + aX 0ij + u00j

This test will analyze whether there is a significant relationship (a) between the school

characteristics in the model (X) and student locus of control (M) (Figure 7).

3. Level 1: Ytij = π0ij + bMij + e tij Level 2: π0ij = β00j + r0ij Level 3: β00j = γ000 + c’X 0ij + u00j

This last mediation test will analyze whether locus of

control (M) has a significant relationship (b) with student occupational aspirations (Y) and

whether the direct relationship (c’) of the school characteristic (X) becomes significantly smaller

in size relative to (c) in equation 1.

Mediation Models Sets

School Effects on Occupational Aspirations via Student Locus of Control (Model Set 1)

The following models have been specified to test the indirect effects of school

characteristics on student occupational aspirations via student locus of control. While the models

are based on the multilevel tests of mediation noted in the previous section, the following models

Figure 6. Path a

Figure 7. Paths b and c'

104

include both the school level characteristics and student level characteristics of interest. Student

characteristics (Sex, SES, Minority Status, Parental Occupation, Curriculum Track) are utilized

primarily as control variables.

As noted in the previous section on mediation testing, Baron and Kenny (1986, p. 1176)

proposed several conditions that needed to be met for a variable to function as a mediator. Based

on their work and updated for use in hierarchical linear models (Zhang et al., 2009, p. 697), the

following paths will be tested: (Path c) whether the school characteristics at Level 3 influence

the occupational aspirations outcome at Level 1 when the mediator (locus of control) is absent;

(Path a) whether variations in the levels of the school characteristics at Level 3 significantly

account for variations in the locus of control, the mediating variable, at Level 1; (Path b) whether

variations in the locus of control mediator at Level 1 significantly accounts for variations in the

occupational aspirations outcome at Level 1; and (Path c’) whether the influence of the school

characteristics at Level 3 on the occupational aspiration outcomes at Level 1 is reduced with the

inclusion of the locus of control mediator at Level 1.

Mediation path c: School effects model for student

occupational aspirations. In this model, the effect of school

characteristics on student occupational aspirations is examined

and student occupational aspirations (OCC30tij) is estimated.

This model takes into account student level characteristics at Level 2 in order to control for their

effects at the starting point (8th grade) and on the change in student occupational aspirations

through 12th grade. School level characteristics from Level 3 are included in order to understand

the extent to which these characteristics predict change in student occupational aspirations from

8th through 12th grades. Level 1 specifies repeated measures of occupational aspirations for

105

students over 8th, 10th, and 12th grades. Level 2 specifies student level characteristics. Level 3

specifies school level characteristics.

Level-1 (OCC30)tij = π0ij + π 1ij(YEAR)tij + e tij Level-2 π0ij = β00j + β01j(FEMALE) + β02j(MINORITY) + β03j(SES) + β04j(PAROCCUP) + r0ij π1ij = β10j + β11j(FEMALE) + β12j(MINORITY) + β13j(SES) + β14j(PAROCCUP) +

β15j(VOCTRACK) + r1ij Level-3 β00j = γ000 + u00j β01j = γ010 β02j = γ020 β03j = γ030 β04j = γ040 β 10j = γ100 + γ101(LOSCHSES) + γ102(CLIMATE) + γ103(ACADPRESS) +

104(NOCOUNSEL) + u10j β 11j = γ 110 β 12j = γ 120 β 13j = γ 130 β 14j = γ 140 β 15j = γ 150 Ytij = average occupational aspirations of a male, non-minority, average SES student

in an academic track with parents of average occupational prestige in a school with at or above average SES, with an average climate, average academic press, and a formal counseling department.

γ000 = average occupation aspirations of an eighth-grade male, non-minority, average SES student with parents of average occupational prestige

γ010 = effect of being female on the average occupational aspirations of an eighth-grade non-minority student who is of average SES with parents of average occupational prestige

γ020 = effect of being a racial minority on the average occupational aspirations of an eighth-grade male student of average SES with parents of average occupational prestige

γ030 = effect of a one unit increase in student SES on the average occupational aspirations of an eighth-grade male, non-minority student with parents of average occupational prestige

γ040 = effect of a one unit increase in occupational prestige of the student’s parents on the average occupational aspirations of an eighth-grade male, non-minority student of average SES

γ100 = average per year change in the occupational aspirations over the course of four years in high school of a male, non-minority student of average SES in an

106

academic track with parents of average occupational prestige in a school with at or above average SES, an average climate, average academic press and a formal counseling department

γ101 = effect of below average school SES on the average per year change in the occupational aspirations of a male, non-minority student of average SES in an academic track with parents of average occupational prestige in a school with average climate, average academic press and a formal counseling department

γ102 = effect of a one unit increase in the positive climate of the school on the average per year change in the occupational aspirations of a male, non-minority student of average SES, in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average academic press, and a formal counseling department

γ103 = effect of a one unit increase in the level of academic press at the school on the average per year change in the occupational aspirations of a male, non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate and a formal counseling department

γ104 = effect of not having access to a formal department of counseling on the average per year change in the occupational aspirations of a male, non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate and average academic press

γ 110 = effect of the student being female on the average per year change in the occupational aspirations of a non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

γ 120 = effect of the student being a racial minority on the average per year change in the occupational aspirations of a male student with average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

γ 130 = effect of a one unit increase in student SES on the average per year change in the occupational aspirations of a male, non-minority student in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

γ 140 = effect of a one unit increase in occupational prestige of the student’s parents on the average per year change in occupational aspirations of a male, non-minority student of average SES in an academic track attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

γ 150 = effect of a student being in a vocational track on the average per year change in the occupational aspirations of a male, non-minority student of average SES with parents of average occupational prestige attending a high school with at or above

107

average SES, average climate, average academic press and a formal counseling department

Mediation path a: School effects model for student

locus of control. In this model, the influence of school level

characteristics on the change in student locus of control is

examined and student locus of control (LOCtij) is estimated.

This model takes into account student level characteristics in order to control for their effects at

the starting point (8th grade) and on the change in locus of control.

Level-1 (LOC)tij = π0ij + π 1ij(YEAR)tij + e tij Level-2 π0ij = β00j + β01j(FEMALE) + β02j(MINORITY) + β03j(SES) + β04j(PAROCCUP) + r0ij π1ij = β10j + β11j(FEMALE) + β12j(MINORITY) + β13j(SES) + β14j(PAROCCUP) +

β15j(VOCTRACK) + r1ij

Level-3 β00j = γ000 + u00j β01j = γ010 β02j = γ020 β03j = γ030 β04j = γ040 β 10j = γ100 + γ101(LOSCHSES) + γ102(CLIMATE) + γ103(ACADPRESS) +

γ104(NOCOUNSEL) + u10j β 11j = γ 110 β 12j = γ 120 β 13j = γ 130 β 14j = γ 140 β 15j = γ 150

Ytij = average locus of control of a male, non-minority, average SES student in an

academic track with parents of average occupational prestige in a school with at or above average SES, with an average climate, average academic press, and a formal counseling department.

γ000 = average locus of control of an eighth-grade male, non-minority, average SES student with parents of average occupational prestige

γ010 = effect of being female on the average locus of control of an eighth-grade non-minority student who is of average SES with parents of average occupational prestige

108

γ020 = effect of being a racial minority on the average locus of control of an eighth-grade male student of average SES with parents of average occupational prestige

γ030 = effect of a one unit increase in student SES on the average locus of control of an eighth-grade male, non-minority student with parents of average occupational prestige

γ040 = effect of a one unit increase in occupational prestige of the student’s parents on the average locus of control of an eighth-grade male, non-minority student of average SES

γ100 = average per year change in the locus of control over the course of four years in high school of a male, non-minority student of average SES in an academic track with parents of average occupational prestige in a school with at or above average SES, an average climate, average academic press and a formal counseling department

γ101 = effect of below average school SES on the average per year change in the locus of control of a male, non-minority student of average SES in an academic track with parents of average occupational prestige in a school with average climate, average academic press and a formal counseling department

γ102 = effect of a one unit increase in the positive climate of the school on the average per year change in the locus of control of a male, non-minority student of average SES, in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average academic press, and a formal counseling department

γ103 = effect of a one unit increase in the level of academic press at the school on the average per year change in the locus of control of a male, non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate and a formal counseling department

γ104 = effect of not having access to a formal department of counseling on the average per year change in the locus of control of a male, non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate and average academic press

γ 110 = effect of the student being female on the average per year change in the locus of control of a non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

γ 120 = effect of the student being a racial minority on the average per year change in the locus of control of a male student with average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

γ 130 = effect of a one unit increase in student SES on the average per year change in the locus of control of a male, non-minority student in an academic track with parents of average occupational prestige attending a high school with at or above

109

average SES, average climate, average academic press and a formal counseling department

γ 140 = effect of a one unit increase in occupational prestige of the student’s parents on the average per year change in the locus of control of a male, non-minority student of average SES in an academic track attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

γ 150 = effect of a student being in a vocational track on the average per year change in the locus of control of a male, non-minority student of average SES with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

Mediation path b: Student locus of control effects on student occupational

aspirations. In this model, the effects of change in

student locus of control on change in student

occupational aspirations is examined (Path b) and

student occupational aspirations (OCC30tij) is estimated.

This model also takes into account several student level

characteristics at Level 2 to control for their effects on the starting point in 8th grade and on the

change in student occupational aspirations and locus of control over time.

Level-1 (OCC30)tij = π0ij + π 1ij(YEAR)tij + π 2ij(LOC)tij + e tij Level-2 π0ij = β00j + β01j(FEMALE) + β02j(MINORITY) + β03j(SES) + β04j(PAROCCUP) + r0ij π1ij = β10j + β11j(FEMALE) + β12j(MINORITY) + β13j(SES) + β14j(PAROCCUP) +

β15j(VOCTRACK) + r1ij π2ij = β20j Level-3 β00j = γ000 + u00j β01j = γ010 β02j = γ020 β03j = γ030 β04j = γ040

110

β 10j = γ100 + u10j β 11j = γ 110 β 12j = γ 120 β 13j = γ 130 β 14j = γ 140 β 15j = γ 150 β 20j = γ200 + u20j Ytij = average occupational aspirations of a male, non-minority, average SES student







γ100 = average per year change in the occupational aspirations over the course of four years in high school of a male, non-minority student of average SES in an academic track with parents of average occupational prestige in a school with at or above average SES, an average climate, average academic press and a formal counseling department




111


γ 150 = effect of a student being in a vocational track on the average per year change in the occupational aspirations of a male, non-minority student of average SES with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

γ200 = average per year change in the occupational aspirations of a male, non-minority student of average SES, in an academic track and with parents of average occupational prestige over the course of four years in high school associated with one unit change in student locus of control

Mediation path c’: School effects model for

student occupational aspirations with mediator. In this

model, the effects of school characteristics on student

occupational aspirations via their effects on student locus

of control are examined and student occupational

aspirations (OCC30tij) are estimated. This model (Path c’) differs from the previous model (Path

c) in that this model includes the mediator at Level 1. It differs from the previous model (Path b)

in that this model includes the school level characteristics at Level 3. With the inclusion of both

the mediator and school level characteristics, this model examines the relationship of school

level characteristics and the effect of change in locus of control on change in occupational

aspirations. As with all the other models, this model takes into account the student level

characteristics at Level 2 in order to control for their effects. Level 1 specifies repeated

measures of locus of control over 8th, 10th, and 12th grades. Level 2 specifies student level

characteristics. Level 3 specifies school level characteristics.

Level-1 (OCC30)tij = π0ij + π 1ij(YEAR)tij + π 2ij(LOC)tij + e tij

112

Level-2 π0ij = β00j + β01j(FEMALE) + β02j(MINORITY) + β03j(SES) + β04j(PAROCCUP) + r0ij π1ij = β10j + β11j(FEMALE) + β12j(MINORITY) + β13j(SES) + β14j(PAROCCUP) +

β15j(VOCTRACK) + r1ij π2ij = β20j Level-3 β00j = γ000 + u00j β01j = γ010 β02j = γ020 β03j = γ030 β04j = γ040 β 10j = γ100 + γ101(LOSCHSES) + γ102(CLIMATE) + γ103(ACADPRESS) +

γ104(NOCOUNSEL) + u10j β 11j = γ 110 β 12j = γ 120 β 13j = γ 130 β 14j = γ 140 β 15j = γ 150 β 20j = γ200 + u20j Ytij = average occupational aspirations of a male, non-minority, average SES student



γ010 = effect of being female on the average occupational aspirations of an eighth-grade non- minority student who is of average SES with parents of average occupational prestige





γ101 = effect of below average school SES on the average per year change in the occupational aspirations of a male, non-minority student of average SES in an

113

academic track with parents of average occupational prestige in a school with average climate, average academic press and a formal counseling department









γ200 = per year change in the occupational aspirations of a male, non-minority student of average SES in an academic track and parents of average occupational prestige attending a high school at or above the average SES, average climate, average

114

academic press and a formal counseling department over the course of four years in high school associated with a one unit change in locus of control

School Effects on Student Occupational Aspirations via Academic Achievement (Model Set 2)

Given that academic achievement may be another avenue through which schools

influence student occupational aspirations, the following models have been specified to test the

indirect effects of school characteristics on student occupational aspirations via student academic

achievement. While the models are based on the multilevel tests of mediation, as in the previous

model set, the following models include both the school level characteristics and student level

characteristics of interest. Student characteristics (Sex, SES, Minority Status, Parental

Occupation, Curriculum Track) are utilized as control variables.

Also consistent with the previous model set, in this model set the following paths will be

tested: (Path c) whether the school characteristics at Level 3 influence the occupational

aspirations outcome at Level 1 when the mediator (achievement) is absent; (Path a) whether

variations in the levels of the school characteristics at Level 3 significantly account for

variations in achievement, the mediating variable, at Level 1; (Path b) whether variations in the

achievement mediator at Level 1 significantly accounts for variations in the occupational

aspirations outcome at Level 1; and (Path c’) whether the influence of the school characteristics

at Level 3 on the occupational aspiration outcomes at Level 1 is reduced with the inclusion of the

achievement mediator at Level 1.

115

Mediation path c: School effects model for student occupational aspirations. This

model was tested in the previous model set. It is reiterated here in order to provide a complete

understanding of the mediation paths in this model set with a different

mediator. As noted in the previous model set, the effect of school

characteristics on student occupational aspirations is examined and

student occupational aspirations (OCC30tij) is estimated. This model

takes into account student level characteristics at Level 2 in order to control for their effects at

the starting point (8th grade) and on the change in student occupational aspirations. School level

characteristics from Level 3 are included in order to understand the extent to which the school

contextual characteristics predict change in student occupational aspirations from 8th through 12th

grades. Level 1 specifies repeated measures of occupational aspirations for students over 8th,

10th, and 12th grades. Level 2 specifies student level characteristics. Level 3 specifies school

level characteristics.

Level-1 (OCC30)tij = π0ij + π 1ij(YEAR)tij + e tij Level-2 π0ij = β00j + β01j(FEMALE) + β02j(MINORITY) + β03j(SES) + β04j(PAROCCUP) + r0ij π1ij = β10j + β11j(FEMALE) + β12j(MINORITY) + β13j(SES) + β14j(PAROCCUP) +


γ104(NOCOUNSEL) + u10j β 11j = γ 110 β 12j = γ 120 β 13j = γ 130

116

β 14j = γ 140 β 15j = γ 150 Ytij = average occupational aspirations of a male, non-minority, average SES student



γ010 = effect of being female on the average occupational aspirations of an eighth grade non-minority student who is of average SES with parents of average occupational prestige









117




γ 140 = effect of a one unit increase in occupational prestige of the student’s parents on the average per year change in the occupational aspirations of a male, non-minority student of average SES in an academic track attending a high school with at or above average SES, average climate, average academic press and a formal counseling department


Mediation path a: School effects model for student academic

achievement. In this model, the influence of school level

characteristics on the change in student academic achievement is

examined and student achievement (ACHIEVEtij) is estimated. This

model takes into account student level characteristics in order to control for their effects at the

starting point in 8th grade and on their change over time.

Level-1 (ACHIEVE)tij = π0ij + π 1ij(YEAR)tij + e tij Level-2 π0ij = β00j + β01j(FEMALE) + β02j(MINORITY) + β03j(SES) + β04j(PAROCCUP) + r0ij π1ij = β10j + β11j(FEMALE) + β12j(MINORITY) + β13j(SES) + β14j(PAROCCUP) +


118

Level-3 β00j = γ000 + u00j β01j = γ010 β02j = γ020 β03j = γ030 β04j = γ040 β 10j = γ100 + γ101(LOSCHSES) + γ102(CLIMATE) + γ103(ACADPRESS) +


Ytij = average achievement of a male, non-minority, average SES student in an


γ000 = average achievement of an eighth-grade male, non-minority, average SES student with parents of average occupational prestige

γ010 = effect of being female on the average achievement of an eighth-grade non-minority student who is of average SES with parents of average occupational prestige

γ020 = effect of being a racial minority on the average achievement of an eighth-grade male student of average SES with parents of average occupational prestige

γ030 = effect of a one unit increase in student SES on the average achievement of an eighth-grade male, non-minority student with parents of average occupational prestige

γ040 = effect of a one unit increase in occupational prestige of the student’s parents on the average achievement of an eighth-grade male, non-minority student of average SES

γ100 = average per year change in the achievement over the course of four years in high school of a male, non-minority student of average SES in an academic track with parents of average occupational prestige in a school with at or above average SES, an average climate, average academic press and a formal counseling department

γ101 = effect of below average school SES on the average per year change in the achievement of a male, non-minority student of average SES in an academic track with parents of average occupational prestige in a school with average climate, average academic press and a formal counseling department

γ102 = effect of a one unit increase in the positive climate of the school on the average per year change in the achievement of a male, non-minority student of average SES, in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average academic press, and a formal counseling department

119

γ103 = effect of a one unit increase in the level of academic press at the school on the average per year change in the achievement of a male, non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate and a formal counseling department

γ104 = effect of not having access to a formal department of counseling on the average per year change in the achievement of a male, non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate and average academic press

γ 110 = effect of the student being female on the average per year change in the achievement of a non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

γ 120 = effect of the student being a racial minority on the average per year change in the achievement of a male student with average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

γ 130 = effect of a one unit increase in student SES on the average per year change in the achievement of a male, non-minority student in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

γ 140 = effect of a one unit increase in occupational prestige of the student’s parents on the average per year change in the achievement of a male, non-minority student of average SES in an academic track attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

γ 150 = effect of a student being in a vocational track on the average per year change in the achievement of a male, non-minority student of average SES with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press and a formal counseling department

Mediation path b: Student academic achievement effects

on student occupational aspirations. In this model, the effects of

change in student academic achievement on change in student

occupational aspirations are examined (Path b) and student

occupational aspirations (OCC30tij) are estimated. This model also takes into account several

120

student level characteristics at Level 2 to control for their effects on the Level 1 variables starting

at 8th grade and on the change over time.

Level-1 (OCC30)tij = π0ij + π 1ij(YEAR)tij + π 2ij(ACHIEVE)tij + e tij Level-2 π0ij = β00j + β01j(SEX) + β02j(MINORITY) + β03j(SES) + β04j(PAROCCUP) + r0ij π1ij = β10j + β11j(SEX) + β12j(MINORITY) + β13j(SES) + β14j(PAROCCUP) +

β15j(VOCTRACK) + r1ij π2ij = β20j Level-3 β00j = γ000 + u00j β01j = γ010 β02j = γ020 β03j = γ030 β04j = γ040 β 10j = γ100 + u10j β 11j = γ 110 β 12j = γ 120 β 13j = γ 130 β 14j = γ 140 β 15j = γ 150 β 20j = γ200 + u20j Ytij = average occupational aspirations of a male, non-minority, average SES student







γ100 = average per year change in the occupational aspirations over the course of four years in high school of a male, non-minority student of average SES in an

121

academic track with parents of average occupational prestige in a school with at or above average SES, an average climate, average academic press and a formal counseling department






γ200 = average per year change in the occupational aspirations of a male, non-minority, average SES student in an academic track with parents of average occupational prestige in a school which is at or above the average school SES, with an average climate, average academic press, and a formal counseling department associated with one unit change in student academic achievement

Mediation path c’: School effects model for student

occupational aspirations with mediator. In this model, the effects

of school characteristics on student occupational aspirations via their

effects on student academic achievement are examined and student

occupational aspirations (OCC30tij) are estimated. This model (Path c’) differs from the previous

model (Path c) in that this model includes the mediator at Level 1. It differs from the previous

122

model (Path b) in that this model includes the school level characteristics at Level 3. With the

inclusion of both the mediator and school level characteristics, this model examines the

relationship of school level characteristics on change in academic achievement and the effect of

change in academic achievement on change in occupational aspirations. As with all the other

models, this model takes into account the student level characteristics at Level 2 in order to

control for their effects. Level 1 specifies repeated measures of locus of control over 8th, 10th ,

and 12th grades. Level 2 specifies student level characteristics. Level 3 specifies school level

characteristics.

Level-1 (OCC30)tij = π0ij + π 1ij(YEAR)tij + π 2ij(ACHIEVE)tij + e tij Level-2 π0ij = β00j + β01j(FEMALE) + β02j(MINORITY) + β03j(SES) + β04j(PAROCCUP) + r0ij π1ij = β10j + β11j(FEMALE) + β12j(MINORITY) + β13j(SES) + β14j(PAROCCUP) +


γ104(NOCOUNSEL) + u10j β 11j = γ 110 β 12j = γ 120 β 13j = γ 130 β 14j = γ 140 β 15j = γ 150 β 20j = γ200 + u20j Ytij = average occupational aspirations of a male, non-minority, average SES student



123












124





School Effects on Student Achievement via Student Locus of Control (Model Set 3)

The primary mediator of interest in this study is student locus of control. However, given

the ties between academic achievement and occupations noted in the theoretical framework, it

was important to analyze the potential role achievement might have in mediating the impact of

schools on student occupational aspirations in the previous model set. Both previous model sets

then set the stage to analyze Hypothesis 3: whether academic achievement may also be indirectly

influenced by school contextual characteristics via student locus of control. The following

models have been specified to test the indirect effects of school characteristics on student

academic achievement via student locus of control. While the models are based on the

multilevel tests of mediation noted in the previous section, the following models include both the

school level characteristics and student level characteristics of interest. Student characteristics

125

(Sex, SES, Minority Status, Parental Occupation, Student Curriculum Track) are utilized as

control variables.

As noted in the previous section on mediation testing, Baron and Kenny (1986, p. 1176)

proposed several conditions that needed to be met for a variable to function as a mediator. Based

on their proposal and updated for use in hierarchical linear models (Zhang et al., 2009, p. 697),

the following paths will be tested: (Path c) whether the school characteristics at Level 3

influence academic achievement outcome at Level 1 when the mediator (locus of control) is

absent; (Path a) whether variations in the levels of the school characteristics at Level 3

significantly account for variations in the locus of control, the mediating variable, at Level 1;

(Path b) whether variations in the locus of control mediator at Level 1 significantly accounts for

variations in the academic achievement outcome at Level 1; and (Path c’) whether the influence

of the school characteristics at Level 3 on the academic achievement outcomes at Level 1 is

reduced with the inclusion of the locus of control mediator at Level 1.

Mediation path c: School effects model for student

academic achievement. This model is the same as that of Path a

in the previous model set. Still, it is included here for the

purposes of providing a complete mediation model. As such, the

effect of school characteristics on student academic achievement is examined and student

academic achievement (ACHIEVEtij) is estimated. This model takes into account student level

characteristics at Level 2 in order to control for their effects at the starting point (8th grade) and

on the change in student occupational aspirations. School level characteristics from Level 3 are

included in order to understand the direct relationship between them without the mediator. Level

1 specifies repeated measures of occupational aspirations for students over 8th, 10th, and 12th

126

grades. Level 2 specifies student level characteristics. Level 3 specifies school level

characteristics.

Level-1 (ACHIEVE)tij = π0ij + π 1ij(YEAR)tij + e tij Level-2 π0ij = β00j + β01j(FEMALE) + β02j(MINORITY) + β03j(SES) + β04j(PAROCCUP) + r0ij π1ij = β10j + β11j(FEMALE) + β12j(MINORITY) + β13j(SES) + β14j(PAROCCUP) +



Ytij = average achievement of a male, non-minority, average SES student in an







γ100 = average per year change in the achievement over the course of four years in high school of a male, non-minority student of average SES in an academic track with parents of average occupational prestige in a school with at or above average

127

SES, an average climate, average academic press and a formal counseling department

γ101 = effect of below average school SES on the average per year change in the achievement of a male, non-minority student of average SES in an academic track with parents of average occupational prestige in a school with average climate, average academic press and a formal counseling department



γ104 = effect of not having access to a formal department of counseling on the average per year change in the achievement of a male, non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate and average academic press

γ 110 = effect of the student being female on the average per year change in the achievement of a non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press and a formal counseling department


γ 130 = effect of a one unit increase in student SES on the average per year change in the achievement of a male, non-minority student in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

γ 140 = effect of a one unit increase in occupational prestige of the student’s parents on the average per year change in the achievement of a male, non-minority student of average SES in an academic track attending a high school with at or above average SES, average climate, average academic press and a formal counseling department


128

Mediation path a: School effects model for student

locus of control. This model is the same as Path a in the first

mediation model set. Still, it is noted here in the interest of

providing a complete mediation model by showing all of the paths being examined. As noted

previously, in this model, the influence of school level characteristics on the change in student

locus of control is examined and student locus of control (LOCtij) is estimated. This model takes

into account student level characteristics in order to control for their effects at the starting point

(8th grade) and on the change in locus of control over time.

Level-1 (LOC)tij = π0ij + π 1ij(YEAR)tij + e tij Level-2 π0ij = β00j + β01j(FEMALE) + β02j(MINORITY) + β03j(SES) + β04j(PAROCCUP) + r0ij π1ij = β10j + β11j(FEMALE) + β12j(MINORITY) + β13j(SES) + β14j(PAROCCUP) +



Ytij = average locus of control of a male, non-minority, average SES student in an


γ000 = average locus of control of an eighth-grade male, non-minority, average SES student with parents of average occupational prestige

129

γ010 = effect of being female on the average locus of control of an eighth-grade non-minority student who is of average SES with parents of average occupational prestige

γ020 = effect of being a racial minority on the average locus of control of an eighth-grade male student of average SES with parents of average occupational prestige

γ030 = effect of a one unit increase in student SES on the average locus of control of an eighth-grade male, non-minority student with parents of average occupational prestige

γ040 = effect of a one unit increase in occupational prestige of the student’s parents on the average locus of control of an eighth-grade male, non-minority student of average SES

γ100 = average per year change in the locus of control over the course of four years in high school of a male, non-minority student of average SES in an academic track with parents of average occupational prestige in a school with at or above average SES, an average climate, average academic press and a formal counseling department

γ101 = effect of below average school SES on the average per year change in the locus of control of a male, non-minority student of average SES in an academic track with parents of average occupational prestige in a school with average climate, average academic press and a formal counseling department

γ102 = effect of a one unit increase in the positive climate of the school on the average per year change in the locus of control of a male, non-minority student of average SES, in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average academic press, and a formal counseling department

γ103 = effect of a one unit increase in the level of academic press at the school on the average per year change in the locus of control of a male, non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate and a formal counseling department

γ104 = effect of not having access to a formal department of counseling on the average per year change in the locus of control of a male, non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate and average academic press

γ 110 = effect of the student being female on the average per year change in the locus of control of a non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

γ 120 = effect of the student being a racial minority on the average per year change in the locus of control of a male student with average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

130

γ 130 = effect of a one unit increase in student SES on the average per year change in the locus of control of a male, non-minority student in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

γ 140 = effect of a one unit increase in occupational prestige of the student’s parents on the average per year change in the locus of control of a male, non-minority student of average SES in an academic track attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

γ 150 = effect of a student being in a vocational track on the average per year change in the locus of control of a male, non-minority student of average SES with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

Mediation path b: Student locus of control effects on student academic

achievement. In this model, the effects of change in student locus

of control on change in student academic achievement is examined

(Path b) and student academic achievement (ACHIEVEtij) is

estimated. This model also takes into account several student level characteristics at Level 2 to

control for their effects on the starting point in 8th grade and on the change in student academic

achievement and locus of control over time.

Level-1 (ACHIEVE)tij = π0ij + π 1ij(YEAR)tij + π 2ij(LOC)tij + e tij Level-2 π0ij = β00j + β01j(FEMALE) + β02j(MINORITY) + β03j(SES) + β04j(PAROCCUP) + r0ij π1ij = β10j + β11j(FEMALE) + β12j(MINORITY) + β13j(SES) + β14j(PAROCCUP) +

β15j(VOCTRACK) + r1ij π2ij = β20j Level-3 β00j = γ000 + u00j β01j = γ010 β02j = γ020 β03j = γ030 β04j = γ040

131

β 10j = γ100 + u10j β 11j = γ 110 β 12j = γ 120 β 13j = γ 130 β 14j = γ 140 β 15j = γ 150 β 20j = γ200 + u20j Ytij = average achievement of a male, non-minority, average SES student in an







γ100 = average per year change in the achievement over the course of four years in high school of a male, non-minority student of average SES in an academic track with parents of average occupational prestige in a school with at or above average SES, an average climate, average academic press, and a formal counseling department

γ 110 = effect of the student being female on the average per year change in the achievement of a non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

γ 120 = effect of the student being a racial minority on the average per year change in the achievement of a male student with average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department


132

γ 140 = effect of a one unit increase in occupational prestige of the student’s parents on the average per year change in the achievement of a male, non-minority student of average SES in an academic track attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department


γ200 = average per year change in achievement of a male, non-minority student of average SES in an academic track and parents of average occupational prestige attending a high school at or above the average SES, average climate, average academic press, and a formal counseling department associated with a one unit change in locus of control

Mediation path c’: School effects model for student

academic achievement with mediator. In this model, the effects of

school characteristics on student academic achievement via their

effects on student locus of control are examined and student

academic achievement (ACHIEVEtij) is estimated. This model (Path c’) differs from the

previous model (Path c) in that this model includes the mediator at Level 1. It differs from the

previous model (Path b) in that this model includes the school level characteristics at Level 3.

With the inclusion of both the mediator and school level characteristics, this model examines the

relationship of school level characteristics on change in locus of control and the effect of change

in locus of control on change in academic achievement. As with all the other models, this model

takes into account the student level characteristics at Level 2 in order to control for their effects.

Level 1 specifies repeated measures of locus of control over 8th, 10th, and 12th grades. Level 2

specifies student level characteristics. Level 3 specifies school level characteristics.

Level-1 (ACHIEVE)tij = π0ij + π 1ij(YEAR)tij + π 2ij(LOC)tij + e tij

133

Level-2 π0ij = β00j + β01j(FEMALE) + β02j(MINORITY) + β03j(SES) + β04j(PAROCCUP) + r0ij π1ij = β10j + β11j(FEMALE) + β12j(MINORITY) + β13j(SES) + β14j(PAROCCUP) +


γ104(NOCOUNSEL) + u10j β 11j = γ 110 β 12j = γ 120 β 13j = γ 130 β 14j = γ 140 β 15j = γ 150 β 20j = γ200 +u20j Ytij = average achievement of a male, non-minority, average SES student in an







γ100 = average per year change in the achievement over the course of four years in high school of a male, non-minority student of average SES in an academic track with parents of average occupational prestige in a school with at or above average SES, an average climate, average academic press, and a formal counseling department

γ101 = effect of below average school SES on the average per year change in the achievement of a male, non-minority student of average SES in an academic track

134

with parents of average occupational prestige in a school with average climate, average academic press and a formal counseling department



γ104 = effect of not having access to a formal department of counseling on the average per year change in the achievement of a male, non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, and average academic press

γ 110 = effect of the student being female on the average per year change in the achievement of a non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department



γ 140 = effect of a one unit increase in occupational prestige of the student’s parents on the average per year change in the achievement of a male, non-minority student of average SES in an academic track attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

γ 150 = effect of a student being in a vocational track on the average per year change in the achievement of a male, non-minority student of average SES with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

γ200 = average per year change in the achievement of a male, non-minority student of average SES in an academic track and parents of average occupational prestige attending a high school at or above the average SES, average climate, average

135

academic press and a formal counseling department associated with a one unit change in locus of control

Full Mediation Model: School Effects on Occupational Aspirations via Locus of Control and Achievement (Model Set 4)

The following model is a culmination of the previous model sets in that it includes both

student locus of control and student academic achievement as a way to test for the indirect

effects of school characteristics on student occupational aspirations. The following model

includes both the school level characteristics and student level characteristics of interest. Student

characteristics (Sex, SES, Minority Status, Parental Occupation, Curriculum Track) are utilized

as control variables.

In this model, the effects of school characteristics on student occupational aspirations via

their effects on student locus of control and academic

achievement are examined and student occupational

aspirations (OCC30tij) are estimated. This model

examines the interaction of the school level characteristics on change in locus of control and the

change in academic achievement and how the change in these variables impacts the effects on

student occupational aspirations over time. Level 1 specifies repeated measures of locus of

control and academic achievement over 8th, 10th, and 12th grades. Level 2 specifies student level

characteristics. Level 3 specifies school level characteristics.

Level-1 (OCC30)tij = π0ij + π 1ij(YEAR)tij + π 2ij(ACHIEVE)tij + π 3ij(LOC)tij + e tij Level-2 π0ij = β00j + β01j(FEMALE) + β02j(MINORITY) + β03j(SES) + β04j(PAROCCUP) + r0ij π1ij = β10j + β11j(FEMALE) + β12j(MINORITY) + β13j(SES) + β14j(PAROCCUP) +


136

π2ij = β20j π3ij = β30j Level-3 β00j = γ000 + u00j β01j = γ010 β02j = γ020 β03j = γ030 β04j = γ040 β 10j = γ100 + γ101(LOSCHSES) + γ102(CLIMATE) + γ103(ACADPRESS) +

γ104(NOCOUNSEL) + u10j β 11j = γ 110 β 12j = γ 120 β 13j = γ 130 β 14j = γ 140 β 15j = γ 150 β 20j = γ200 + u20j β 30j = γ300 + u30j Ytij = average occupational aspirations of a male, non-minority, average SES student









137


γ103 = effect of a one unit increase in the level of academic press at the school on the average per year change in the occupational aspirations of a male, non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, and a formal counseling department

γ104 = effect of not having access to a formal department of counseling on the average per year change in the occupational aspirations of a male, non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, and average academic press

γ 110 = effect of the student being female on the average per year change in the occupational aspirations of a non-minority student of average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

γ 120 = effect of the student being a racial minority on the average per year change in the occupational aspirations of a male student with average SES in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

γ 130 = effect of a one unit increase in student SES on the average per year change in the occupational aspirations of a male, non-minority student in an academic track with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

γ 140 = effect of a one unit increase in occupational prestige of the student’s parents on the average per year change in occupational aspirations of a male, non-minority student of average SES in an academic track attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

γ 150 = effect of a student being in a vocational track on the average per year change in the occupational aspirations of a male, non-minority student of average SES with parents of average occupational prestige attending a high school with at or above average SES, average climate, average academic press, and a formal counseling department

γ200 = average per year change in the occupational aspirations of a male, non-minority, average SES student in an academic track with parents of average occupational prestige in a school which is at or above the average school SES, with an average climate, average academic press, and a formal counseling department associated with one unit change in locus of control

138


Significance and Effect Sizes of Mediation

Testing whether student locus of control mediates the relationship between school

characteristics and student occupational aspirations requires additional tests once the models

above are estimated. According to Baron and Kenny (1986), when paths a and b are controlled,

a previously significant relationship between the independent and dependent variables is no

longer significant, with the strongest demonstration of mediation occurring when Path c is zero.

The quantification of mediation effects can be performed in different ways, two of which

include the product-of-coefficients method (the product of paths a and paths b = ab) or the

difference-in-coefficients method (path c – c’). Both methods are considered valid and roughly

equivalent in single-level models, however, within multiple levels these methods can produce

different values which may be interpreted differently (Krull & MacKinnon, 1999; MacKinnon,

2008), but this is if multiple mediators are present (Zhang et al., 2009, p. 699). In the case of this

analysis, both methods will be utilized to understand whether mediation is present.

If mediation has been established, then it is necessary to understand whether the level of

mediation is significant. To understand the significance levels of the mediation, MacKinnon

(2008) noted that the mediated effect must be significantly different from zero and suggests that

one way to test for this is to estimate a confidence interval and assess whether zero is included in

the confidence interval. If zero is outside of the confidence interval, then the mediated effect is

139

statistically significant, or if the value of the ratio exceeds 1.96, then the mediated effect is

significantly different from zero at p < .05 (MacKinnon, 2008, p. 53).

MacKinnon (2008) also pointed out that significance tests do not provide information on

the size and meaningfulness of a mediation effect. In order to understand the size and

meaningfulness, MacKinnon suggests that while ideal effect size measures for mediation are yet

to be resolved, analyzing the partial correlation of the mediated paths is one way currently in use.

To analyze the partial correlation for Path c’ (rYX.M which is the correlation between X and Y

partialled for M) and Path b (rYM.X which is the correlation between Y and M, partialled for X),

the following formulas are suggested (MacKinnon, 2008, p. 86):

rYX.M = 𝑟𝑋𝑌 − 𝑟𝑀𝑌 𝑟𝑋𝑀

��1−𝑟𝑀𝑌2 �(1−𝑟𝑋𝑀

2 ) (Path c’)

rYM.X = 𝑟𝑀𝑌 − 𝑟𝑋𝑌 𝑟𝑋𝑀

��1−𝑟𝑋𝑌2 �(1−𝑟𝑋𝑀

2 ) (Path b)

Taken together, the tests of significance, effect size, and meaningfulness will allow a more

precise understanding of the meaningfulness of the mediation effects between schools, locus of

control, and occupational aspirations.

140

CHAPTER V

ANALYSIS

An analysis of the results from each of the models reveals the relationship between

school contextual characteristics, the growth of student occupational aspirations, student locus of

control, and academic achievement. Following is a discussion of these results and their

relationship to the hypotheses and framework of the study.

Dependent Variables and Growth Over Time

Initial ANOVA models were run for the three dependent variables (student occupational

aspirations, student locus of control, and student academic achievement) to test the first

hypothesis and determine whether there was growth in these variables over time (see Table 15).

Table 15

Descriptive Statistics for ANOVA’s


Locus of Control

Academic Achievement

M SD Max Min M SD Max Min M SD Max Min Time of Repeated Measure

8th 54 9 76 0 3 .51 4 1 47 8 67 25 10th 63 14 76 0 3 .49 4 1 52 9 73 25 12th 63 14 76 0 3 .52 4 1 55 10 81 26

Results reveal that for student occupational aspirations, there is significant and positive

growth in occupational aspirations over time (π1ij = 2.37, p < .001) and that while there is

141

variation in the average growth in occupational aspirations across schools (u10 = .21, χ2 =

1830.76 with a standard deviation of .46 and p < .001), with plausible mean growth values for

95% of schools between 1.47 and 3.27, the magnitude of that variation in mean growth in

occupational aspirations across schools is small, at less than one-fifth of one standard deviation.

Additionally, the annual growth rate of student occupational aspirations across schools is highly

and positively correlated (Τβ =.959) with initial student occupational aspirations in 8th grade.

While student aspirations tend to show significant and positive growth across schools, Figure 8

provides a visual representation of a small, random sample of individual students and their

occupational aspirations at 8th, 10th, and 12th grades, revealing that individual student

occupational aspirations change over time, with some students having decreases in aspirations.

Since occupational aspirations are a primary focus of this study, it is helpful to interpret

these results in relationship to the Nakao-Treas (1994) occupational prestige scale. As noted in

the descriptive statistics, the grand mean occupational aspiration of students in 8th grade is 54.

This score is roughly equivalent to the occupational prestige of jobs such as kindergarten teacher,

TV or radio announcer, construction supervisor, or general administrator. The grand mean

occupational aspirations of students in 10th and 12th grade increase to a prestige score of 63,

roughly the equivalent to occupations such as educational administrator, automotive plant

manager, economist, nuclear engineer, author, banker or meteorologist. While the mean reflects

the average aspirations across schools for 8th, 10th, and 12th grades, the results of the

unconditional model indicate that individual student aspirational scores increase each year by

2.37. An example of how this growth might be interpreted is illustrated by considering the

142

Figure 8. Sample of Change in Occupational Aspirations from 8th to 12th Grades. Year: 0 = 8th grade, 2 = 10th grade, 4 = 12th grade

occupational prestige of the student with an average score (54) in 8th grade. If that student’s

score grew by the average of 2.37 prestige points each of the following years through 12th grade,

then in the 10th grade their score would be 58.74 (54+2.37+2.37), roughly equivalent to the

prestige of occupations such as an advertising executive, public relations manager, mechanical

engineer, journalist, fire inspector, detective, or secret service agent. In 12th grade, their score

would be 63.48 (58.74+2.37+2.37), roughly equivalent to the grand mean occupational prestige

score of 63 (corresponding occupations for this score are noted above). Obviously, given the

variation in initial scores across schools in 8th grade and their variation in growth, not all students

will have initial occupational aspirations in 8th grade beginning at the mean of 54, nor will all

students aspirational scores grow by the grand mean average of 2.37 each year—although the

average score growth across 95% of schools will be between 1.47 and 3.27.

Results for student locus of control reveal that the average growth across schools in locus

of control is both negative and nonsignificant (π1ij = -.004, p > .05). Still, an analysis of the chi-

36.10

46.55

57.00

67.45

77.90

Occ

upat

iona

l Asp

irat

ions

-0. 1 2 3 4

YEAR

143

squared statistics show that there is variation in the growth in locus of control across schools (u10

= .001, χ2 = 2334.63 with a standard deviation of .03 and p <.001), with 20% of the variation in

change being between schools and the magnitude of that variation being small (less than one

fourth of one standard deviation) and 95% of schools having variations in changes in locus of

control between -.067 and .059, allowing for a decrease, or change from an internal locus of

control and toward a more external locus of control over time. While average growth in locus of

control across schools is nonsignificant, Figure 9 is a small sample of individual students that

provides a visual representation of individual student change in locus of control from 8th to 10th

to 12th grades.

Figure 9. Sample of Change in Locus of Control From 8th to 12th Grades. Year: 0 = 8th grade, 2 = 10th grade, 4 = 12th grade

-0. 1 2 3 40.87

1.59

2.30

3.02

3.73

YEAR

Locu

s of

Con

trol

144

Finally, results for student academic achievement reveal that the grand mean growth for

academic achievement is positive and significant (π1ij = 2.21, p > .001). An analysis of the

variation in achievement shows that there is variation in the average growth (u10 = .181, χ2 =

3136.28 with a standard deviation of .43 and p < .001), with the magnitude of variation in

growth being small (less than one-fourth of one standard deviation), and plausible values for

variation in change for 95% of schools between 1.38 and 3.04. Correlations confirm a positive

relationship (Τβ = .094) between mean academic achievement in 8th grade and growth in mean

achievement. While the results provide information on the growth in academic achievement

across schools, Figure 10 provides a visual representation of individual student

Figure 10. Sample of Change in Academic Achievement From 8th to 12th Grades. Year: 0 = 8th grade, 2 = 10th grade, 4 = 12th grade

-0. 1 2 3 432.71

43.11

53.51

63.91

74.31

YEAR

Acad

emic

Ach

ieve

men

t

145

achievement across 8th, 10th, and 12th grades taken from a small, random sample of individual

students.

The results for all ANOVA models are summarized in Table 16. The subsequent

hypotheses themselves are built upon the theory that there would be change in the dependent

variable. Indeed, even as there is slight variation in average change in locus of control across

schools, the lack of significant change in locus of control over time is unexpected.

Table 16 Results for Individual ANOVA Models

Results for Individual ANOVA Models

Coefficient for

Growth (γ10)

Magnitude of

Variation Variance

(U10)

Chi-Squared

(χ2) SD Df

Correlations to initial

status (Τβ)

Occupational Aspirations 2.37* (1.47, 3.27) .21* 1830.76 .46 1314 .959

Locus of Control -.004†

(-.067, .059) .001* 2334.63 .03 1314 -.723

Academic Achievement 2.21* (1.38, 3.04) .18* 3136.28 .43 1314 .094

* significant at p < .001; † nonsignificant.

The ANOVA for occupational aspirations, locus of control, and academic achievement

suggest that student occupational aspirations and academic achievement grow while locus of

control remains significantly unchanged over time. However, these models do not show results

in relation to each other as they were run independent of any predictor variables (i.e., the

occupational aspirations model did not include locus of control or academic achievement as

146

predictors) and they do not control for the effects of student level variables such as race, gender,

SES, parental occupation, or curriculum track and as such, any conclusions drawn prior to

running more complete models would be premature.

Unlike the ANOVA models discussed above, the unconditional growth model (Model 4)

examines growth in occupational aspirations in relation to growth in student locus of control and

student academic achievement and provides a baseline understanding of the relationship between

the three variables. Without any student or school level variables, the results of this model reveal

that the average occupational aspirations of 8th grade students across all schools is 57.11 (p <

.001) and the average growth in occupational aspirations across schools is 1.57 prestige points

per year and is significant at p < .001. A one unit increase in student locus of control (toward a

more internalized locus) across schools is associated with the average growth in occupational

aspirations by 1.35 prestige points per year, significant at p < .001, and a one unit increase in

student academic achievement across schools is positively and significantly (p < .001) associated

with the average growth in student occupational aspirations across schools by .36 prestige points

per year (see Table 17).

Results of this model suggest that change in both academic achievement and locus of

control positively and significantly influence the growth in student occupational aspirations. It is

helpful to understand the size of the effects of growth in locus of control and academic

achievement on growth in occupational aspirations. To determine effect size, the following

equation has been used: (coefficient for growth x standard deviation/standard deviation of

occupational aspirations) with the understanding that closer to 1, the larger the effect size. The

effect size of growth in locus of control on growth in occupational aspirations is .05 and the

147

Table 17 Results for Model 4: Full Growth Model

Results for Model 4: Full Growth Model

Average Growth in OCC30 (γ) Effect Size

Occupational Aspirations (OCC30) 1.57* --

Associated with a 1 unit increase in Locus of

Control (LOC) 1.35* .05

Associated with a 1 unit increase in Academic

Achievement (ACHIEVE) .36* .27 *significant at p < .001.

effect size of growth in achievement on growth in occupational aspirations is .27. While initial

conclusions regarding student occupational aspirations, locus of control and achievement cannot

be drawn from an analysis of the ANOVA models, the unconditional model provides greater

detail into the relationship among the three variables, suggesting that growth in achievement and

locus of control are significantly and positively associated with growth in occupational

aspirations.

Mediation Models

The first hypothesis in this study proposes that positive school characteristics (positive

climate, counseling support, at or above average school SES, and more academic press) will be

associated with gains in occupational aspirations over time. This hypothesis will be tested as

148

part of the mediation model sets outlined below. Table 18 outlines the results of all the

mediation models.

School Effects on Occupational Aspirations via Student Locus of Control (Model Set 1)

Hypothesis 2 proposes that student locus of control will mediate the

relationship between school contextual variables and growth in

student occupational aspirations. Paths in this model set have been

specified to test this hypothesis. Analysis of the paths in this model set provides information on

the indirect effects of school characteristics including school climate, counseling support, school

SES, and academic press on student occupational aspirations via student locus of control and

whether hypothesis 2 can be supported. Each path was modeled independently in order to

ascertain the mediation effects.

Path c. As noted previously, this path tests whether there is a significant direct

relationship between the school contextual characteristics and the growth in student occupational

aspirations. The school characteristics analyzed in this study did not have a significant effect on

growth in student occupational aspirations. In fact, all of the four variables included were

nonsignificant, even though there was significant and positive growth (γ = 2.39, p <.001) in

student occupational aspirations over time. While all the school characteristics had

nonsignificant effects, the coefficients for school SES (γ101 = -.113, p = .609) and school climate

(γ102 = -.126, p = .281) were negative, while academic press (γ103 = .175, p = .110) and

counseling support (γ104 = .092, p = .627) has small, positive effects.

149

Table 18

Results of all Mediation Model Sets

Note. SCH = school characteristics, OCC = student occupational aspirations, LOC = student locus of control, ACH = student academic achievement. *significant at p < .001, **significant at p < .05, †nonsignificant

Growth in Outcome Variable

(γ100)

Effect of Mediating Variable

(γ200)

Between School

Variance in Growth

(U10) χ2 SD df Model Set 1 SCHLOC OCC

Path C SCHOCC 2.39*

.09* 1752.33 .30 1310

Path A SCHLOC -.003†

.001* 2283.99 .02 1310

Path B LOCOCC 2.39*

2.05* .11* 1880.09 .33 1302

Path C’ SCHLOCOCC 2.39* 2.05* .11* 1878.56 .32 1300 Model Set 2 SCHACH OCC

Path C SCHOCC 2.39* .09* 1752.33 .30 1310 Path A SCHACH 2.21*

.14* 2857.56 .37 1310

Path B ACHOCC 1.64* .34* .22* 2133.22 .47 1314 Path C’ SCHACHOCC 1.65* .34* .22* 2131.01 .47 1310 Model Set 3 SCHLOC ACH Path C SCHACH 2.21* .14* 2857.56 .37 1310 Path A SCHLOC -.003† .00* 2283.99 .02 1310 Path B LOCACH 2.21* .67* .14* 2775.97 .38 1304 Path C’ SCHLOCACH 2.21* .67* .14* 2742.03 .37 1300 Full Model SCHLOCACH OCC 1.69*

1.16* LOC .33* ACH .24† 1108.35 .49 1185

150

Path a. This study hypothesized that school contextual variables might have an effect on

occupational aspirations via a mediator, specifically student locus of control. Path a specifically

tests hypothesis 2a, and an analysis of the test on Path a provides an understanding of a potential

mediating effect and whether there is a significant relationship between the school contextual

characteristics in the model and growth in student locus of control (toward an internal locus of

control).

Results show that school contextual characteristics were nonsignificant in the change in

student locus of control, which in itself produced a nonsignificant result (γ = -.003, p = .330).

Since results of Path a reveal that school context characteristics do not influence student locus of

control over time, hypothesis 2a is not supported.

Path b. Complete mediation analysis requires an examination of the final two paths, b

and c’. Path b tests hypothesis 2b and examines whether student locus of control influences

student occupational aspirations. Results of Path b indicate that growth in student locus of

control over time has a positive and significant effect on growth in student occupational

aspirations (γ = 2.05, p < .001), which supports hypothesis 2b.

Path c. When full or partial mediation exists, the direct effects of school contextual

characteristics found in the original Path c should be smaller (partial mediation) or fully reduced

to zero (full mediation) in Path c’. The direct relationship between school contextual variables

and student occupational aspirations was tested in Path c, which showed the direct effect of the

school contextual variables on growth in student occupational aspirations to be nonsignificant.

Similarly, Path c’ reveals that school contextual effects on student occupational aspirations

remains nonsignificant, even with the inclusion of student locus of control. Table 19 summarizes

the results of model set 1.

151

Table 19 School Contextual Effects for Model Set 1

Model Set 1 SCHLOCOCC

Path C SCHOCC

Path A SCHLOC

Path B LOCOCC

Path C’ SCHLOCOCC

Growth in Dependent Variable 2.39* -.003† 2.40* 2.40* Effect of LOC on Growth in OCC 2.05* 2.05* School SES -.113† -.008† -.084† Counseling Support .092† .005† .070† Positive Climate -.126† -.003† -.127† Academic Press .175† .002† .177†

Note. SCH = school characteristics, OCC = student occupational aspirations, LOC = student locus of control. *significant at p <.001; †nonsignificant.

Mediation analysis. With results from all of the paths in this model set, the steps

outlined by Baron and Kenny (1986) were utilized to analyze whether there is a mediation effect.

The first step is to establish that the independent variables have an effect on the dependent

variable, which requires an analysis of Path c. MacKinnon (2008) noted that the purpose of this

initial step is to ensure there is an effect to mediate. In the case of this model set, none of the

school contextual variables were shown to have a significant effect on student occupational

aspirations. Since that was not the case, and Path c was nonsignificant, it might seem that any

further mediation analysis would be unnecessary. Still, MacKinnon pointed out that the necessity

of establishing an initial effect between the independent and dependent variables prior to any

additional steps is controversial since “it is possible that the relation between the independent

variable and the dependent variable may be nonsignificant, yet there can still be substantial

mediation” (p. 68), and a relationship that is not seen initially in an analysis of Path c may exist,

152

but through a mediating variable. MacKinnon goes on to suggest that the most important paths

for much of mediation in the social sciences are paths a and b, and analysis of the relationships in

these two paths are often the only steps required to establish mediation. Given the results of Path

c, this is the expectation for model set 1 and analysis of potential mediation for this model set

moves to the next step.

The second step in mediation analysis is to ensure that the independent variables are

related to the mediating variable, which is Path a. In the case of model set 1, this requires

analyzing the relationship between school contextual variables and student locus of control. This

relationship also proved to be nonsignificant.

In the third step, the mediator must affect the dependent variable while controlling for the

independent variable (Path b). In model set 1, Path b is the relationship between student locus of

control and student occupational aspirations while controlling for the school contextual variables.

For model set 1, Path b shows a positive, significant relationship.

The final step in the mediation analysis is to assess whether there is a mediation effect,

either full or partial, which means that either the direct relationship between the school

contextual variables are now nonsignificant where they were previously significant (full

mediation) or that Path c’ is now less than Path c, but not necessarily reduced to zero (partial

mediation). Recalling MacKinnon’s (2008) comment that paths a and b are often the most

important in determining whether mediation may exist, it is important to note that while Path b in

this model set is significant, Path a is not. MacKinnon cautions that

generally, the test of b̂ (Path B) is not sufficient to demonstrate a mediation effect because a researcher will typically require the relation between the independent variable and the mediator, the â (Path A) regression coefficient, to be statistically significant. (p. 69)

153

Therefore, there is no mediated relationship and analysis of model set 1 concludes with the

assessment that school contextual characteristics do not influence the development of student

occupational aspirations either directly or indirectly through student locus of control, since there

is not a significant relationship between the school contextual variables and growth in student

locus of control. Hypotheses 2a and 2c are not supported. However, growth in student locus of

control (toward a more internal locus) is associated with growth in student occupational

aspirations, and therefore hypothesis 2b is supported.

Results in relation to hypotheses. In this model set, hypothesis 2a was tested in which it

was expected that positive school contextual characteristics would be associated with a change

toward an increased locus of control.

Results of Path a revealed that school climate, school SES, school counseling support,

and academic press do not influence student locus of control.

Path b tested hypothesis 2b in which it was expected that the more internal a student’s

locus of control over time, the more likely the student would be to have gains in occupational

aspirations over time. The results of testing this pathway revealed that growth toward a more

internal locus of control does have a positive influence on occupational aspirations. If a

student’s locus of control increases (moving toward a greater internal locus) by one unit, then the

occupational aspirations of that student also tend to increase by 2.05 prestige points.

Path c’ tested whether student locus of control mediated the relationship between school

contextual characteristics and occupational aspirations. The results reveal that since school SES,

climate, academic press, and counseling support do not influence student locus of control, then

locus of control cannot mediate the relationship between school SES, climate, academic press,

and counseling support and student occupational aspirations.

154

In this study, hypothesis 2a is unsupported, hypotheses 2b is supported and 2c is

unsupported.

School Effects on Student Occupational Aspirations via Student Academic Achievement (Model Set 2)

Hypothesis 3 proposes that student academic achievement will

mediate the relationship between school context variables and

student occupational aspirations. Paths in model set 2 have been

specified to test this hypothesis and analysis of the paths in this model set provides information

on the indirect effects of school characteristics including school climate, counseling support,

school SES, and academic press on student occupational aspirations via student academic

achievement, since academic achievement may be another avenue through which schools

influence student occupational aspirations. Each path was modeled independently in order to

ascertain the mediation effects.

Path c . As in Model Set 1, this path tests whether there is a significant direct

relationship between the school contextual characteristics and the growth in student occupational

aspirations. As with model set 1, the school characteristics analyzed did not have a significant

effect on growth in student occupational aspirations.

Path a. As previously discussed, this study hypothesizes that school contextual variables

might have an effect on student occupational aspirations via a mediator. While student locus of

control was an initial focal point for potential mediation, literature on the intersections between

school contextual characteristics, academic achievement and occupational aspirations brought

achievement to the fore as another potential mediator through which school context might

influence student occupational aspirations. Path a specifically tests hypothesis 3a and an analysis

155

of the test on Path a provides information on a potential mediating effect and whether there is a

significant relationship between the school contextual characteristics in the model and growth in

student academic achievement.

The coefficient for growth in student academic achievement is positive and significant (γ

= 2.21, p < .001). In addition, all school contextual characteristics are shown to be

nonsignificant in the growth of student academic achievement except for academic press, which

has a small, positive and significant impact on growth (γ103= .184, p < .05). Since results of Path

a reveal that academic press is one of the school contextual characteristics that significantly

influence student academic achievement over time, hypothesis 3a is partially supported.

Path b. Path b tests hypothesis 3b and examines whether student academic achievement

influences student occupational aspirations. Results of Path b indicate that growth in student

academic achievement over time has a positive and significant effect (γ = 1.64, p < .001) on

growth in student occupational aspirations, increasing growth in aspirations by 1.64 prestige

points, which supports hypothesis 3b.

Path c’. As described in model set 1, when full or partial mediation exists, the direct

effects of school contextual characteristics found in the original Path c should be smaller (partial

mediation) or fully reduced to zero (full mediation) in Path c’. In this model, as in model set 1,

the relationship between the school contextual characteristics and student occupational

aspirations is nonsignificant. Table 20 summarizes the results of model set 2.

Mediation analysis. Following the steps outlined in the previous model set, potential

mediation in model set 2 will be analyzed. While the first step in Baron and Kenny’s mediation

analysis (1986) is to establish that the independent variables have an effect on the dependent

variable (Path c) to ensure there is an effect to mediate, the results of Path c in this model set are

156

Table 20 School Contextual Effects for Model Set 2

Note. SCH = school characteristics, OCC = student occupational aspirations, ACH = student academic achievement. *significant at p <.001, †nonsignificant the same as Path c in the previous model set in that none of the school contextual variables are

shown to have a significant effect on student occupational aspirations. Since prior work in

mediation establishes that the product of coefficients method (the products of paths a and paths b

= ab) is also a valid method for analyzing mediation (Cohen & Cohen, 1983; MacKinnon, 2008;

MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002), model set 2 will be analyzed for

mediation effects, despite the nonsignificant relationship revealed by the analysis using the

differences in coefficients method (paths c-c’).



analyzing the relationship between school contextual variables and student academic

achievement. With the exception of the academic press, these relationships also proved to be

Model Set 2 SCHACHOCC

Path C SCHOCC

Path A SCHACH

Path B ACHOCC

Path C’ SCHACHOCC

Growth in Dependent Variable 2.39* 2.21* 1.64* 1.65* Effect of ACH on Growth in OCC .34* .34* School SES -.113† .024† .041† Counseling Support .092† -.074† .076† Positive Climate -.126† -.082† -.082† Academic Press .175† .184* .085†

157

nonsignificant. Since academic press has a significant and positive effect (γ = .184, p < .05) on

academic achievement, Path a is a significant pathway.


independent variable (Path b). In model set 2, Path b is the relationship between student

academic achievement and student occupational aspirations while controlling for the school

contextual variables. For model set 2, Path b shows a significant and positive relationship (γ =

.34, p < .001).

The final step in the mediation analysis is to assess whether there is a mediation effect,

either full or partial, which means that in an examination of paths c and c’ that either the direct

relationship between the school contextual variables are now nonsignificant where they were

previously significant (full mediation) or that Path c’ is now less than Path c, but not necessarily

reduced to zero (partial mediation). In an analysis of mediation via paths a and b it is important

to note that Path a has one contextual variable with a positive and significant effect on the

mediator and Path b has a significant and positive effect on the dependent variable. Unlike

model set 1, the analysis of model set 2 concludes with the assessment that, using the product of

coefficients method (paths ab), there is a mediated relationship: the school contextual

characteristic of academic press does influence the development of student occupational

aspirations indirectly through student academic achievement. The mediated effect of ab is

(.184)(.34) = .063. This is a significant effect (p < .001), which was validated through the

calculation of Sobel’s z test and the subsequent analysis of the confidence interval around the

mediated effect (Iacobucci, 2008; MacKinnon, 2008). Given the significant mediated effect of

academic press on academic achievement and the effect of academic achievement on student

occupational aspirations, hypothesis 3 is supported.

158

Results in relation to hypotheses. In model set 2, hypothesis 3a was tested in which it

was expected that positive school contextual characteristics would be associated with gains in

academic achievement over time. Results of Path a revealed that school climate, school SES,

and school counseling support do not influence student achievement, but a one unit increase in

the encouragement of students by teachers to achieve academically (academic press) increased

academic achievement of students by .184 unit, providing partial support for hypothesis 3a.

Path b tested hypothesis 3b in which it was expected that gains in academic achievement

would be associated with gains in occupational aspirations over time. The results of testing this

pathway revealed that when a student has a one unit increase in academic achievement, they are

also likely to have an increase in their aspirations of .34 prestige points, providing support for

hypothesis 3b.

Path c’ tested whether student academic achievement mediated the relationship between

school contextual characteristics and occupational aspirations, which was the foundation for

hypothesis 3c. Given the mediation results in which a one unit increase in academic press

increases a student’s occupational aspirations by .063 prestige points, hypothesis 3c is partially

supported. .

For this study, hypothesis 3a is partially supported, 3b is supported, and 3c is partially

supported.

School Effects on Student Achievement via Student Locus of Control (Model Set 3)

Hypothesis 4 proposes that student locus of control will mediate the

relationship between school context variables and growth in student

159

achievement. The paths in this model set have been specified to test this hypothesis and analysis

of the paths in this model set provides information on the indirect effects of school

characteristics including school climate, counseling support, school SES, and academic press on

student academic achievement via student locus of control, as a way to understand whether

student locus of control might mediate school contextual influence on student academic

achievement since there is subsequent interest in whether achievement might influence student

occupational aspirations (see model set 2). Each path was modeled independently in order to

ascertain the mediation effects, with the following results.

Path c. As noted previously in Path a of model set 2, this path analyzes the relationship

between school contextual variables and growth in student academic achievement. In this path,

the coefficient for growth in student academic achievement is positive and significant (γ = 2.21,

p <.001). Academic press is the only school contextual characteristic that has a significant

effect, with a small, positive and significant impact on growth in academic achievement (γ103=

.184, p < .05).

Path a. Path a in this model set is the same at Path a in model set 1 and reveals whether

there is a significant relationship between the school contextual characteristics in the model and

student locus of control. Results show that school contextual characteristics were nonsignificant

in the change in student locus of control, which in itself produced a nonsignificant result (γ = -

.003, p = .311, n.s.).

Path b. Path b tests hypothesis 4a and analyzes whether locus of control influences

student academic achievement. Results of Path b indicate that growth in locus of control toward

a more internal locus is positively and significantly associated with gains in student academic

achievement (γ = .667, p <.001), which supports hypotheses 4a.

160

Path c’. With mediation analysis using the differences in coefficients method (path c-c’),

the direct effects of school contextual characteristics found in the original Path c should be

smaller (partial mediation) or fully reduced to zero (full mediation) in Path c’. In the case of

model set 3, academic press is the only characteristic that has a significant effect on achievement

and this effect does not decrease from Path c to Path c’, but rather increases slightly from γ =

.184, p < .05 to γ = .190, p < .05. Table 21 summarizes the results of model set 3.

Table 21

School Contextual Effects for Model Set 3

Model Set 3 SCHLOCACH

Path C SCHACH

Path A SCHLOC

Path B LOCACH

Path C’ SCHLOCACH

Growth in Dependent Variable 2.21* -.003† 2.21* 2.21* Effect of LOC on Growth in ACH .67* .67* School SES .024† -.008† .018† Counseling Support -.074† .005† -.077† Positive Climate -.082† -.003† -.082† Academic Press .184* .002† .190*

Note. SCH = school characteristics, LOC = student locus of control, ACH = student academic achievement. *significant at p <.001, †nonsignificant

Mediation analysis. Following the steps outlined in the previous model sets, potential

mediation in model set 3 will be analyzed. The first step in Baron and Kenny’s mediation

analysis (1986) is to establish that the independent variables have an effect on the dependent

variable (Path c) to ensure there is an effect to mediate. The results of Path c in this model set

reveal that one school contextual variable, academic press, is shown to have a significant effect

161

on the dependent variable, student academic achievement. As such, mediation analysis of model

set 3 continues.



analyzing the relationship between school contextual variables and student locus of control. This

relationship proved to be nonsignificant.


independent variable (Path b). In model set 3, Path b is the relationship between student locus of

control and student academic achievement while controlling for the school contextual variables.

For model set 3, Path b shows a positive, significant, relationship.

The final step in the mediation analysis is to assess whether there is a mediation effect.

An analysis of the differences in Path c and Path c’ shows that there is not a reduction in the

effects with the inclusion of the mediating variable, and thus the difference in coefficients

analysis does not support a mediating effect. While paths a and b are often the most important in

determining whether mediation may exist, it is important to note that although Path b in this

model set is significant, Path a is not. Therefore, the product of coefficients analysis also reveals

a lack of a mediating effect. Thus, the analysis of model set 3 concludes with the assessment that

there is no mediation effect. School contextual characteristics do not influence the development

of student academic achievement indirectly through student locus of control, since there is not a

significant relationship between the school contextual variables and student locus of control.

Hypothesis 4b is not supported.

Results in relation to hypotheses. In model set 3, hypotheses 4a and 4b were tested. In

hypothesis 4a it was expected that the more internal a student’s locus of control, the more likely

162

a student would be to have gains in academic achievement over time. Results of Path b revealed

that if a student had a one unit increase toward a more internal locus of control, the student

would also see an increase in academic achievement of .67 units, supporting hypothesis 4a.

Hypothesis 4b expected that student locus of control would mediate the relationship

between school SES, climate, academic press and counseling support and student academic

achievement. Since none of these characteristics influence a student’s sense of control, there was

no mediated pathway and therefore hypothesis 4b was not supported.

For this study, hypothesis 4a is supported and hypotheses 4b is unsupported

Full Mediation Model: School Effects on Occupational Aspirations via Locus of Control and Achievement (Model Set 4)

The full mediation model was specified to test whether,

with the inclusion of both mediators (student locus of

control and student academic achievement), school

contextual characteristics would significantly influence growth in student occupational

aspirations, as noted in hypothesis 1.

In this model, growth in student locus of control positively and significantly (γ20 = 1.16,

p < .001) influences the growth of student occupational aspirations (γ10 = 1.69, p < .001), as

does growth in student academic achievement (γ30 = .325, p < .001).

Mediation analysis. An analysis of the school contextual characteristics reveals them to

be nonsignificant across the board, as noted in the Table 22 below. While aligned with the

reported results found in previous model sets, this result is disappointing in the context of the

study which theorizes a mediated pathway through which schools might influence student

occupational aspirations. Taken together with the mediation analysis of the three previous model

163

Table 22

School Contextual Effects on Dependent Variables From Full Model

Full Model SCHLOCACH OCC


Academic Achievement

Locus of Control

Growth 1.69* .33* 1.16* School SES .050† Counseling Support .065† Positive Climate -.087† Academic Press .091†

Note. SCH = school characteristics, LOC = student locus of control, ACH = student academic achievement, OCC = student occupational aspirations. *significant at p <.001, †nonsignificant

Student Level Characteristics

Student SES, race, gender, academic track, and parental occupational status are

characteristics included in order to analyze their role in relation to the dependent variables. As

noted in the discussion of the theoretical model in Chapter III, the student curriculum track

variable is of some interest in that it was initially thought to have potential influence on student

occupational aspirations, locus of control, and academic achievement. Analysis of the results of

the student level characteristics reveal that curriculum track is the student level variable with the

largest, significant effect in relation to the outcome variables of occupational aspirations, locus

of control, and achievement. Student gender is the second largest, significant effect. SES is

significant only in relation to achievement while parental occupation is significant only in

relation to locus of control and achievement, and race is nonsignificant.

164

Previous literature has discussed the potential role of curriculum tracking on perceptions

of self and subsequent development of aspirations development. The results of the current study

are in keeping with the literature, revealing that students who reported being in a vocational track

showed a negative influence on growth of internal locus of control (γ = -.019, p < .001). The

results of the model sets in this study also suggest that being in a vocational track has significant,

negative influence on growth of occupational aspirations (γ150 = -1.88, p < .001) and academic

achievement (γ150 = -.507, p < .001).

Gender also has an influence on locus of control, occupational aspiration, and

achievement. A brief look at the study results on the effects of student gender shows that girls

have higher occupational aspirations (γ110 = .837, p < .001), a more internal locus of control (γ110 =

.038, p < .001) but also tend to have also tend to have slightly lower levels of academic

achievement (γ110 = -.257, p < .001).

Parental occupational prestige does not have a significant influence on occupational

aspirations, although it has a very small but negative influence on student locus of control (γ140 = -

.001, p < .001) and student academic achievement (γ140 = -.004, p < .05). Current results reveal

that SES is not significantly related to locus of control or occupational aspirations, but does have

an effect on academic achievement (γ130 = .197, p < .001), which is in line with previous research.

Table 23 summarizes the effects of student characteristics.

Summary of Findings

The initial hypothesis in the study is that positive school contextual characteristics will be

associated with gains in occupational aspirations over time. An overview of the hypotheses and

165

Table 23 Effects of Student Level Characteristics

Note. SCH = school characteristics, LOC = student locus of control, OCC = student occupational aspirations, ACH = student academic achievement. *significant at p < .001, **significant at p < .05, †nonsignificant

Sex

(γ110)

Race (γ120)

SES (γ130)

Parent Occupation

(γ140) Curriculum Track

(γ150) Model Set 1 SCHLOCOCC Path C: SCHOCC .837* .209† .217† .006† -1.88* Path A: SCHLOC .038* .010† .003† -.001* -.019* Path B: LOCOCC .771* .198† .249** .007† -1.84* Path C’: SCHLOCOCC .773* .206† .231† .006† -1.84* Model Set 2 SCHACHOCC Path C: SCHOCC .837* .209† .217† .006† -1.88* Path A: SCHACH -.257* .036† .197* -.004** -.507* Path B: ACHOCC .926* .147† .184† .005† -1.52* Path C’: SCHACHOCC .928* .140† .180† .005† -1.52* Model Set 3 SCHLOCACH Path C: SCHACH -.257* .036† .197* -.004** -.507* Path A: SCHLOC .038* .010† .003† -.001* -.019* Path B: LOCACH -.287* .034† .202* .005† -.510* Path C’: SCHLOCACH -.282* .031† .190* .005† -.508* Full Model: SCHLOCACHOCC .896* .150† .201† .004† -1.51*

166

the subsequent results found in Table 24 suggest that given the measures utilized, only one of the

contextual characteristics of schools specified in this study is associated with growth in student

occupational aspirations—but not through student locus of control, as initially envisioned.

Rather than school contextual characteristics influencing change toward an internal locus of

control over time which would, in turn, influence change in student occupational aspirations, the

study reveals that there is not an association between school contextual characteristics and

change in student locus of control. This result eliminates locus of control as a mediator through

which school characteristics could influence change in student occupational aspirations and

leaves hypotheses 2a and 2c unsupported.

Alternatively, academic press is one of the school characteristics in the study that is

associated with change in student occupational aspirations via student academic achievement,

which mediates the relationship as proposed in hypothesis 3c. So, while this study did not find

locus of control to be a mediating variable, a relationship between school context and

occupational aspirations was found to be mediated, alternatively, through academic achievement

which provides support for hypothesis 1. Other hypotheses supported by the study were the

relationships between the mediators and the dependent variables, unrelated to school contextual

characteristics (specifically, 2b, 3b, and 4a). Still, the data in this study did not find associations

between three of the four school contextual variables specified (school climate, school SES, and

counseling support) and the development of student occupational aspirations over time, even as

the models above show growth in occupational aspirations over time despite that lack of

association.

167

Table 24 Summary of Hypotheses

Hypothesis Path Supported or Unsupported

1

Positive school contextual characteristics will be associated with gains in occupational aspirations over time SCHOCC Partially Supported

2a

Positive school contextual characteristics will be associated with change toward an internal locus of control over time SCHLOC Unsupported

2b

The more internal a student’s locus of control over time, the more likely the student is to have gains in occupational aspirations over time LOCOCC Supported

2c

Student locus of control partially mediates the relationship between school contextual characteristics and occupational aspirations

SCHLOC OCC Unsupported

3a

Positive school contextual characteristics will be associated with gains in academic achievement over time SCHACH Partially Supported

3b

Gains in academic achievement over time will be associated with increases in occupational aspirations over time ACHOCC Supported

3c

Student academic achievement partially mediates the relationship between school contextual characteristics and occupational aspirations

SCHACH OCC Partially Supported

4a

The more internal a student’s locus of control, the more likely the student is to have gains in academic achievement over time LOCACH Supported

4b

Student locus of control will mediate the relationship between school contextual characteristics and student academic achievement

SCHLOC ACH Unsupported

Note. SCH = school contextual characteristics, OCC = student occupational aspirations, LOC = student locus of control, ACH = student academic achievement.

168

The results of this study are interesting in that previous research provides ample

background support for testing the relationships in this study and given the research, one would

expect to find stronger relationships between the contextual variables and student occupational

aspirations. Given that significant relationships do not exist between the bulk of specified

contextual environmental variables of schools and the key student variables in this study that

influence student outcomes is, in itself, of interest even if unexpected. Discussion of these

findings in relationship to the framework for the study, previous and future research and

education policy is the focus of the following chapter.

169

CHAPTER VI

CONTEXTUALIZATION OF RESULTS

The Problem of Misalignment and Social Learning Theory

Part of the title for this study is “learning what I want to be when I grow up” which

encompasses the broad purpose underlying the examination of the specific contextual

characteristics of schools in relation to students and their sense of personal control over their

lives and the development of their aspirations. Children spend the bulk of 13 formative years

walking the hallways of school buildings, sitting in classrooms in front of teachers, and studying

subjects from a carefully selected curriculum. Children become students and schools are their

daily ecosystem. In this ecosystem, children are not only learning about academic subjects, but

they are learning about themselves and they are developing their aspirations for the future: they

are learning what they want to be when they grow up. The research on student outcomes—what

happens after students leave the school system—reveals a disturbing trend: students are growing

up and failing to become what they want to be. They are aspiring to college educations and yet

failing to graduate with college degrees in record numbers. They are aspiring to specific

professions, yet failing to develop successful pathways to those professions. While in school,

they may have been graded on core subjects such as mathematics and science, but they were not

assessed on how they were assimilating this information for their own futures, how their

education was influencing their aspirations and their choices toward those aspirations. Even as

previous research shows that increasingly students are making choices that are misaligned with

their desired outcomes, this information comes too late for those students being studied.

Research reveals retrospectively the information on alignment that students need

170

contemporaneously, information that they need infused into their daily educational ecosystem.

As the percentage of students with misaligned aspirations outnumbers those with aligned

aspirations (Rosenbaum, 2001), there is more at risk than individual earnings and economic

status. National economic stability, technological and scientific progress, and global positioning

are all at risk. Our nation relies on our educational system and the subsequent outcomes of

students to ensure our future place in the world. As such, understanding the disparity between

those students who aspire to specific outcomes and those students who attain their desired

outcomes is important to policymakers, educators, parents, and to the students themselves.

The subtitle of this study is “the contextual effects of schools on student locus of control

and the development of student occupational aspirations,” because it is the influence of

environment, and specifically the schools where children spend so many formative years that is

the context of focus. Rotter’s social learning theory (1954) in which the interaction between

people and their environment motivates behavior created the framework for studying the

contextual effects of schools on students. Grounded in previous research and theory, the current

study identified specific school level variables that, hypothetically, could or should be

influencing students and then reanalyzed their potential direct effects on the development of

student occupational aspirations while it also introduced mediated mechanisms (student locus of

control and student academic achievement) through which the school variables might show an

indirect effect on the development of student occupational aspirations. School socioeconomic

status, whether a school has a positive climate, whether there is encouragement to achieve

academically, and whether there is counseling support within a school were the four variables

that made up the school contextual characteristics studied. A brief synopsis of Chapter V might

lead one to believe that the current study is, on balance, unimportant in that only one of the four

171

school contextual variables showed an impact on the development of student occupational

aspirations. However, such a simple overview may miss the potential complexity of what these

results may indicate.

School socioeconomic status (SES) was hypothesized to impact student development of

occupational aspirations for several reasons outlined in the literature: lower scores on school SES

may signal a structural reinscription of inequalities that exist for students outside of school (i.e.,

in their neighborhoods) and may also indicate a deficit in resources such as counseling programs

or high quality teachers available within the school. Similarly, higher SES schools may indicate

an abundance of resources available to students such as higher quality teachers and counseling

supports (Flowers et al., 2003; Gamoran, 1996; Hanson, 1994; Kelly & Lee, 2001; McWhirter et

al., 2000; Mortimer et al., 2002; Orfield, 1997; Plucker, 1998), all of which may influence the

development of student occupational aspirations via student locus of control or student academic

achievement. The finding in this study that school SES has no significant influence on the

development of student occupational aspirations whether school SES is high or low might

suggest that students may be disconnected from the context in which they are developing

occupational aspirations over time. If, as the literature suggests, lower SES schools may have

fewer resources such as counseling available to students, those students may be unaware of the

lack of resources such as counseling support that they may need to seek out in order to inform

their occupational aspirations. In the case of schools with higher SES, it may suggest that even

though there may be resources and information available to them, students may be unaware that

they may need to avail themselves of those resources and information in order to foster aligned

aspirations. In both cases, the context of school SES is not associated with the development of

the occupational aspirations of students.

172

Similarly, the climate in the school is important for fostering achievement and internal

locus of control (Rumberger, 1995) and may be reflected in the commitment of teachers to their

classrooms (V. E. Lee & Bryk, 1989) and encouragement from teachers (A. M. Ryan & Patrick,

2001). It was hypothesized that a positive school climate, as measured by structured classroom

activity and teacher morale, would influence the development of student locus of control and

subsequently student occupational aspirations. However, the current study finds in opposition of

Rumberger’s (1995) work, that school climate has no significant effect on the development of

student locus of control or student achievement or thus indirectly on the development of student

occupational aspirations. Low morale and unstructured classrooms have no more of an effect on

student’s sense of control or achievement than high teacher morale and highly structured

classrooms. As noted earlier with school SES, the lack of significant effect of the contextual

climate of school may suggest that students may be disengaged from the daily context in which

they are learning and developing, and such disengagement may lead to disconnections between

learning and aspirations and subsequent outcomes.

Since educational and occupational aspirations are so closely intertwined and

achievement and perceptions of control are related, academic press was included as a variable

that might influence the development of student locus of control, achievement and subsequent

occupational aspirations. Measured by whether teachers encourage students to achieve and

whether students are expected to do homework, the results of this study show a significant but

small indirect effect of academic press on occupational aspirations. While previous literature

suggests that encouragement to achieve and to do well academically should matter a great deal to

students, this study clearly shows that while it matters, its influence appears to be small.

Students may not be placing as much stock in the expectations of their teachers, or the

173

encouragement to achieve when developing their aspirations as educators and policymakers

might hope.

Finally, one might expect that the availability of information and support for one’s

occupational aspirations would influence students, perhaps giving them a greater sense of control

over their decisions toward occupational pathways, or assisting them in aligning their aspirations

with their desired outcomes. As such, the availability of formal guidance counseling was

measured. Once again, the availability of school resources was not a significant influence on

student’s locus of control, achievement or subsequent development of occupational aspirations,

signaling that students may be making decisions either without or in spite of the input and

support of informational resources.

For each of the variables included in the study there is ample research pointing to the

importance of these variables in the development of student locus of control, achievement and

occupational aspirations. Still, even as previous research may have noted their significance, the

current study found no significant influence of three of the four school contextual variables either

directly or indirectly on student occupational aspirations, even as student locus of control and

student academic achievement both influence the development of student occupational

aspirations over time. Only academic press showed a small influence on aspirational

development. Taken together, the answers to the research questions posed in this study seem not

to support its theoretical framework built on social learning theory: the interaction between

students and their school environment is not motivating their behavior. Is it possible that social

learning theory is wrong, that students are immune to any effects of the environment in which

they spend so much time? How else does one explain the results of this study?

174

One explanation of these results may come from viewing them from a different

perspective by asking the following question: Are these school characteristics designed to

influence students? With this question in mind, the school characteristics can then be viewed

from a program evaluation perspective in which the goals of the contextual characteristics are

important to the subsequent outcomes (Rossi, Lipsey, & Freeman, 2004). In the case of

counseling support, the results seem to indicate a disconnection between students and the need

for information as either a venue to increase their locus of control or develop appropriate

pathways for their educational and occupational aspirations. For example, if the goal of

counseling departments is to provide students with appropriate information toward the

development of aligned pathways, but counseling support has no significant effect on the

development of occupational aspirations either directly or indirectly through an association with

student locus of control or student academic achievement, then one might call into question

whether the goals associated with counseling departments are aligned with the desired student

outcomes.

A program evaluation perspective spurs the question of whether schools have goals for

contextual characteristics such as climate and counseling in relation to their influence on student

outcomes. Based on the literature, it was expected that the school contextual variables in this

study would have an influence on student occupational aspirations either directly or indirectly.

Social learning theory supports the idea that context influences behavior. Yet, with results

revealing a lack of association between the school context variables and the outcome variable of

student occupational aspirations, one might question whether the goals schools may have in

relation to school context influencing student behavior and outcomes are aligned in achieving

those outcomes. If social learning theory is correct and context does motivate behavior, and the

175

goals of schools for how their context motivates behavior and influences student outcomes such

as occupational aspirations do not align with actions and programs that are in place to attain

those outcomes, then is it possible that such institutional misalignment could be the contextual

characteristic that is influencing students. Is it possible that the results of this study are signaling

that social learning is taking place, but perhaps what students are learning that is motivating their

behavior is the misalignment between educational goals and outcomes that schools themselves

may be modeling, rather than alignment and connection? If the research on student perceptions

of contextual supports or barriers can affect levels of engagement (Daniels & Araposthasis,

2005; Klem & Connell, 2004) which in turn affects level of achievement (Montalvo et al., 2007;

Wehlage et al., 1989) and sets the groundwork for student aspirations and future attainment, then

student perceptions of misaligned contextual supports may be the reason for the lack of influence

of school context on their locus of control, achievement, and development of occupational

aspirations. The logic follows that if students perceive that the desired goals of schools for

students are disconnected from actions and actual student outcomes, then students may perceive

school as irrelevant both to themselves and their futures, fueling misalignment of their own.

Policy Implications and Recommendations

While students may think otherwise, the cited literature on the school contextual

variables included in this study is clear: school context has mattered and should continue to

matter to student outcomes, especially in light of the results of the current study. At a time when

students may see school as irrelevant to their sense of control, achievement, and future

aspirations, when students may be disregarding the desire of teachers and schools for positive

student outcomes, and not availing themselves of the resources and programs which may be

176

available to them, it is even more critical that schools work to positively influence students and

their subsequent outcomes. With the relevance of schools clearly established in the literature, the

issue becomes one of clearly communicating that relevance to students and having that

communication reflected in their school environment.

The idea of relating school to the world of work has been hotly debated in academic

circles since the formation of the public school system in the US (Dewey, 1900, 1897/1959;

Rury, 2002), with educators often arguing for a separation of the two in order to give all children

regardless of social standing an opportunity at an unstratified education and subsequent

opportunities beyond what a stratified, work-oriented education might provide. While the

intentions to provide the best education to all students have been positive, such a separation

between school and the world of work has had inadvertent ramifications in that school is being

perceived by students as irrelevant to their occupational aspirations and attainment and students

are leaving schools with misaligned aspirations which they are failing to attain. According to

Rosenbaum (2001), any effective change in perceptions about the relevance of schools must have

a dual focus that addresses the internal motivation of students and includes external incentives

beyond a “college only” choice. Unfortunately, linkages between academic effort and

occupational outcomes remain fuzzy to students. They need to see that their performance in

school matters to their outcomes in order to see school as relevant, whether they go to college or

go into the workforce. Stern (2009) suggested that relevance of school to students’ lives

involves connecting their educational experiences to their future civic responsibilities and

earning a living. Schneider and Stevenson (1999) discussed the importance of students

connecting their academic efforts with their own desired outcomes via planning and taking into

account the steps, motivation, and resources necessary to ensure their plans come to fruition.

177

Dewey (1900) first gave voice to these ideals at the beginning of the last century during a similar

era of change, recognizing the need for education to reflect the changing culture. Dewey felt that

schools shouldn’t remain couched in a “mediaeval conception of learning” (p. 26), narrowly

focused on the memorization of symbols and recitation of facts. Rather, education should be

experiential, engaging not only a student’s brain but their imagination and abilities, their

tendencies “to make, to do, to create and to produce” (Dewey, 1900, p. 26) which would

contribute to the betterment of their society. While only a sampling, this literature emphasizes

the importance of making education relevant to students by connecting the work they do in

schools with their desired outcomes once they leave school. Based on the findings in this study,

the broad recommendations found in the literature, and the acknowledgement of the complexity

that accompanies educational policy initiatives, the following are considerations for how the

contextual characteristics of schools might reflect their relevance to student aspirations and

future attainment.

1. Create and offer multiple pathways to work and college. As indicated by the

results of this study, curriculum track influences student occupational aspirations,

locus of control and achievement. Rather than continuing with curriculum tracks that

separate and stigmatize career and vocational education, Stern (2009) proposes

reimagining curriculum as a combination of career and technical education (CTE)

with core academic curriculum as a way to prepare students for both college and

economic self-sufficiency. Stern cites longitudinal research that reveals students who

had a combined program had more success in postsecondary education and work than

students who chose a college or career prep track alone. Ideal implementation of this

concept would require integrating the value of career and college into the daily

178

ecosystem of school so that the educational context would reflect the educational

goals. For example, students would be encouraged to discuss their aspirations with

school counselors, who would be encouraged to discuss and provide information and

resources on a broader range of options for students’ educational and occupational

pathways, not just the pathway to college. Academic press would include

encouraging students to connect their achievement academically with educational and

occupational aspirations and attainment after high school. This consideration is not a

new one and there are versions of such initiatives running in various school districts

across the country. While it is still too early for researchers to know whether such

programs are having the desired effect in enhancing student outcomes long term,

initial results have been positive (Stern, 2009, p. 227).

2. Incorporate relevancy of academic work to occupational work into the

curriculum. As previously noted, occupations are less clearly delineated and less

stable than they were in the past, with people having multiple jobs throughout their

working lives. In addition, students have only a vague understanding of what

different jobs entail, the types of skills and education necessary, and how one would

enter a job or profession (Csikszentmihalyi & Schneider, 2000). While students are

seeking relevancy in what they are learning, educators should also desire students to

successfully be able to apply the knowledge and skills learned across their academic

careers throughout their working lives once they leave the realm of academia. To that

end, classroom curriculum should include explicit and clear connections for how what

is being learned is applicable and necessary beyond the classroom. It should also aim

to expand the scope of the students’ occupational knowledge (i.e., occupational

179

sectors, educational pathways, and requisites, necessary soft skills) as they move

through their educational careers, fueling connections between their academic work

and occupational aspirations and assisting in aligning their aspirations with successful

attainment. Incorporation could be as basic as classroom discussion regarding the

relevancy of current subject matter to the realm of occupations or might include

service learning, mentoring, apprenticeship, or other types of projects in which

students have the opportunity to test the relevancy of their learning and solidify

transferrable knowledge by putting that knowledge into action and anchoring

understanding in experience (Bransford, Brown, & Cocking, 1999; Eyler & Giles,

1999).

Another way to incorporate relevancy between academic and occupational

work into the curriculum is to incorporate explicit teaching modules on the alignment

and development of student aspirational pathways. Since research shows that

students develop aspirations in spite of their context, that there is a majority of

students having misaligned aspirations, and that educators are concerned about

student misalignment of aspirations and outcomes, one way to alleviate misalignment

is for schools to do what they do best: educate students on this critical issue.

Csikszentmihalyi and Schneider (2000) highlighted the need for learning

environments to be relevant to student outcomes by including clear information,

experiences, connections, and supports to enhance student internal motivation and

external actions toward their occupational aspirations. Educating students about

aspirational alignment and the ramifications of misalignment, and helping them

actively develop aligned pathways may help alleviate misalignment of student

180

aspirations, but it also signals to students the relevancy of education to their futures.

Once again, the integration of relevancy into the daily ecosystem of the school signals

to students that schools not only desire positive student outcomes beyond high school

but have aligned their educational goals toward that end. By signaling alignment in

their context, schools may be subtly influencing student aspirational alignment as

well.

3. Assess student understanding of educational relevancy. In conjunction with

incorporating relevancy into the curriculum, students should be assessed and coached

(Bransford, 1993) on their ongoing understanding of academic relevancy throughout

their academic careers since “the purpose of learning is to use what is learned” (Eyler

& Giles, 1999, p. 66). One way students perceive what is important in the classroom

is through the process of testing as a way to measure learning. While assessment

does not ensure learning, research suggests that if students are involved in goals of

mastery, learning, and challenge and believe that a task is important, they will engage

in more metacognitive activity (Pintrich & De Groot, 1990). Assessing the ability to

understand the transfer of knowledge and skills from the classroom to the working

world would signal to students that schools believe in the relevancy of education

beyond the classroom, and that the ability to transfer knowledge to that realm is

important. For teachers, assessing student knowledge of relevance between

educations and occupations would underscore the importance of teaching that

relevance and would help ensure that it would garner an ongoing classroom focus. In

addition, such an assessment would provide educators and researchers with a way to

gauge the extent to which students have aligned occupational and educational

181

aspirations. Finally, such assessment would be both an actualization and signal of

alignment between what schools profess they want for students and the actions they

take to help ensure student outcomes, both of which might influence the alignment of

student occupational aspirations.

4. Develop institutional linkages to the working world. Csikszentmihalyi &

Schneider (2000) noted that in order for students to have successful outcomes, they

not only need clear information about the relevancy and connections between school

and work, they need to be able to “learn and practice job relevant behavior” (p. 215).

Rosenbaum’s work (2001) suggested that clear linkages between schools and

organizations allow for students to gain important information that may help them

align their aspirations, including work culture, requisite educational knowledge, and

expectations of necessary soft skills. The school to work opportunities act (STWOA,

1994) tried to incorporate a similar initiative, but failed due to teachers being unsure

of how to go about creating relationships between their classrooms and the

organizations. One consideration for developing such linkages would be to prioritize

this initiative at the administrative level to determine a plan of action for the school

community and initiate the relationships, and then engage counselors and teachers to

work together on implementation. Programs could consist of service learning

projects, mentorships, work study, apprenticeships or other such programs to provide

students with exposure to the working world and expand their understanding of

connections between school and the world of work. To be effective, such linkages

should be established by schools that are committed to developing programs to

enhance the aspirational alignment of students via relevant connections between

182

school and work. These linkages should signal to students the school level

commitment to communicating the relevancy of school to their futures and assisting

them in aligning their aspirations toward attainment.

5. Restructure and reincorporate counseling resources. Rather than perpetuate a

“college for all” approach which does little to address the “forgotten half”

(Rosenbaum, 2001), a new counseling initiative would recognize that fully

implemented guidance programs have positive relationships with both achievement

and occupational development (Borders & Drury, 1992) by restructuring counseling

to be an integral part of educating students toward aligned outcomes rather than

leaving it as an optional—and disregarded—resource for students. Throughout the

academic life of a student, the student would be exposed to and interact with

counselors. Counselors would become a necessary part of the classroom and

curriculum to help bridge student understanding between academics and occupational

outcomes, expanding the student scope of knowledge about the world of work. For

example, counselors might work with teachers in exploring how an academic subject

matter might relate to occupational competence and achievement. They might assist

in teaching sections on pathway planning and development that would be built into a

more aligned curriculum. In addition, counselors could be responsible for

coordinating linkage programs between schools and organizations to garner

community support for positive student outcomes, perhaps assist teachers with

developing service learning or other learning opportunities related to the local

community and the world of work, and work with students in obtaining exposure to

occupational interests in relation to their aspirational pathways. Rather than reducing

183

the role of counselors to a limited, late-stage resource for students attending college,

this initiative would expand the role of counselors to help both students and schools

bridge academic and occupational aspirations with aligned outcomes throughout the

entire educational lifecycle of students.

While the above policy considerations do not explicitly address all of the school

contextual variables in the current study, the connections are inherent in the implementation of

such initiatives. By developing a culture of alignment in which students learn not only what they

should know and be able to do for their specific grade level but how that knowledge is valuable

for the future, schools infuse their climate with signals to students that they are invested in their

aspirations (positive climate). Counseling departments would be focused on aspirational

alignment and relevancy education (counseling support), while teachers would be encouraging

students to know and demonstrate that relevancy (academic press). In developing linkages to

organizations, communities would become active participants in the lives and aspirations of

students as well, perhaps alleviating stigmas that students in lower socioeconomic areas might

otherwise feel regarding their aspirations (school SES). Policy considerations like those noted

above not only offer suggestions for bridging the chasm of perceived irrelevancy between school

and the world of work, but help schools provide context and education around the critical issue

of aspirational alignment and attainment, and educating students on how to leave the classroom

better prepared to attain their goals. Policy initiatives such as these assist not only in realigning

the aspirations of students with desired outcomes, but assist in realigning the aspirational

pathways of schools with their desired outcomes for students. While versions of some of these

suggestions are beginning to take place in school districts across the country, there is still much

work to be done at the policy level to ensure that students understand the importance of schools

184

to their future aspirations and make use of the resources and information available to them in

their daily context of school.

Limitations of the Study and Opportunities for Future Research

Pintrich (2000) noted that “the social and academic domains need to be more fully

integrated in our models of learning” (p. 221), and suggests that a more fully formed

understanding of the influence of context on the individual within that context is important for

the future of education. It is also important for educational research. While the current study

makes an important contribution to understanding how school contextual characteristics might

indirectly influence the development of student occupational aspirations via student locus of

control and academic achievement, the study has limitations. Both the limitations and the

interesting results of the study suggest directions for future research in the area of the contextual

influence of schools on student development.

Overall, the NELS:88 provided a robust dataset with which to address the questions

posed in this study and, overall, was well suited for the desired analysis. However, there were a

few limitations of the NELS:88 dataset in terms of the variables available for use within the

study. First, the results of the study show growth in student occupational aspirations over time.

However, the study was limited in the range of occupations provided to the students in the

sample. The small range was reflected as a tighter grouping of scores on Nakao-Treas (1994)

occupational prestige scale, and a more fully realized list of potential occupations might provide

future researchers with a broader distribution of scores and a more detailed understanding of

growth in aspirations over time. Second, in terms of the school contextual variables, an

understanding of the depth of counseling support available to students was desired. However,

185

due to how the survey question was worded in one of the waves it was not possible to ascertain

how many full-time staff were dedicated to supporting the counseling needs of students. As

such, counseling support was limited to knowing only whether or not the school had a formal

counseling department. Again, greater knowledge of the depth of counseling support might

provide a more intimate understanding of the value schools place on having counseling resources

available to students. Third, availability of additional variables regarding school context such as

whether variables included in the study such as school climate were tied to specific goals for

student outcomes, whether and how students were encouraged to utilize counseling resources,

and whether and how students were encouraged to explore occupational areas would have

allowed a more comprehensive test of the proposed theories within this study.

The results of the study offer some important directions for future researchers. In

examining any potential indirect influence of school context on student occupational aspirations

via mediators, this study contributes the answer to the other half of the question on whether the

context of schools is influencing the development of student occupational aspirations, since the

results confirm that that school SES, school climate, and counseling support do not directly or

indirectly influence aspirations, and the results suggest that these school context variables have

no significant indirect influence on the development of student occupational aspirations via locus

of control or academic achievement over time. Based on the premise of social learning theory—

that interaction between an individual and their environment motivates behavior—and the results

of this study which suggest the lack of association between school context and student

development of locus of control, achievement, and occupational aspirations, it becomes an

imperative of future research to examine whether these contextual variables should be

influencing students, in what ways, to what extent, and what other contextual variables could be

186

influencing student development of occupational aspirations. Additionally, given the importance

of aligned aspirations and relevance of education highlighted by the results of this study, a

deeper examination of the influence of the alignment of school actions with desired student

outcomes on student alignment of aspirations and perceptions of relevance is warranted.

Final Words

The literature is clear regarding the importance of student occupational aspirations.

Similarly, research underscores the importance of school in relation to student outcomes. What

the current study highlights is the continued misalignment of student aspirations in relationship

to school, as school resources, climate, and encouragement have no significant effect on the

development of their locus of control, academic achievement, and occupational aspirations.

While previous research has shown that key school contextual variables have no direct influence

on the development of student occupational aspirations, the current study reveals that key school

contextual variables have no indirect influence on student occupational aspirations either. In

essence, students may be disregarding key elements of school as irrelevant to their futures. But

students alone cannot bear the blame for their misaligned aspirations. Schools may be failing to

communicate the relevancy of education to the future of what students want to become even as

influencing positive student outcomes are core aspirations of public education. Thus, schools

may be misaligned in their own aspirational pathways in relation to student outcomes. A school

context of misalignment may breed the misaligned aspirations of students—and while the school

context should matter in that it is positively influencing aligned aspirations, this study suggests

that it might not matter. As an important educator and educational reformer, John Dewey

believed passionately that education is a social process, and

187

the school is simply that form of community life in which all those agencies are concentrated that will be most effective in bringing the child to share in the inherited resources of the race and to use his own powers for social ends. (1897, p.7)

For the sake of the future, educators and policy makers must ensure that the educational context

in which students are developing matters to their development, that school is not only perceived

as relevant but is so grounded in relevance that students cannot disregard it.

188

APPENDIX

LITERATURE MATRIX

Section Key Points Literature Cited Definitions of Gottfredson, 1981; Hansen &

MacIntire, 1989; Rojewski, 1996; Paul, 1997

Formation of: Status Attainment Theory

Sewell, 1969; Haller & Portes, 1973; Mau & Bikos, 2000; Lee & Rojewski, 2009

Educational & Occupational Aspirations and their Importance

A modern phenomenon Lee & DeVore, 1975; Shanahan, 2000; Csikszentmihalyi & Schneider, 2000; Choy, 2002; Goyette, 2008; Goyette and Mullen, 2006; Mortimer 2008; Modell, 1989; Trusty, Niles & Carney, 2005; Savickas, 2009

Tied to the underpinning of American social structure

Sorokin, 1927; Blau & Duncan, 1967; Sewell, 1969; Sewell, Haller & Ohlendorf, 1970; Haller & Portes, 1973; Haller & Woefel, 1972

Influence on future attainment Strong, 1953; Sewell, 1969; Rojewski 1996; Inoue, 1999; Blau & Duncan, 1967; Gottfredson, 1981; Marjoribanks, 1985; Schoon & Parsons, 2002; Paul, 1997

Closely intertwined Woefel, 1972; Inoue, 1999; Sorokin, 1927

●more education equates to higher status

Gamoran, 2000; Wong, 1997

●education has impact on occupation long term

Jencks, 1979; O'Rand, 2000; Schoon & Parsons, 2002; Rosenbaum, 2001

●sociological implications: labor market and global

competitiveness

Becker, 1964

189

Linkages between Occupational & Educational Aspirations

Academic achievement links aspirations and outcomes

Gamoran, 1996; Helwig, 2004; NCES, 2006; Rindfuss 1999; Rosenbaum 2001; Woefel, 1972

●enhances future occupational opportunities

Parcel & Dufur, 2001

●weak performance and academic attainment

predicts lower earnings

Miller & Rosenbaum, 1997

●Jobs, wages function of education

Johnson & Mortimer, 2002; Csikszentmihalyi & Schneider, 2000; Mau, 1995

Occupational aspirations have direct effects on educational aspirations

●fueled by desire for higher status

Csikszentmihalyi & Schneider, 2000; Inoue, 1999

●occupation strong determinant of wealth

Johnson & Mortimer, 2002

●public associations between education and

preparation for work

Orfield, 1997; Steinberg, 1992; Okagaki, 2001; Csikszentmihalyi & Schneider, 2000; Goyette, 2008

Constrained opportunities impact both education and occupation

●at an early age Lee & Rojewski, 2009 due to background inequalities

(i.e. SES, ability) Bourdieu, 1973; Kerkchoff, 1976; McClelland, 1990; Sewell, 1969

Linkages between Occupational & Educational Aspirations (cont'd)

●due to school characteristics Lee, Bryk & Smith, 1993; Oakes, Gamoran & Page, 1992; Lee & Bryk, 1989; Gamoran, 2000; Gamoran & Berends, 1987; Kerckhoff, 2000; Schneider & Stevenson, 1999; Johnson & Mortimer, 2002

●low educational aspirations linked to low occupational

attainment

Inoue, 1999; Blau & Duncan 1967; Rojewski, 2005

Individual assessment impacts outcomes

Sewell, 1969

●early achievement enhances self selection for future

opportunity

Alwin & Otto, 1993; Mortimer, 2000

●self-assessment reinforced, modified over time - regulates

access, engagement, opportunities

O'Rand, 2000; Lent, 1996; Kerckhoff, 1974; Rindfuss, 1999; Helwig, 2004; Lee & Rojewski, 2009; Kenny 2003

190

Function of schools in relation to society, preparation

Hallinan, 2000; Savickas, 2009; Csikszentmihalyi & Schneider, 2000; Rosenbaum, 2001; Schneider & Stevenson, 1999; Paul, 1997

Connections between school and work

Paul, 1997; Rosenbaum, 2001; Kerckhoff, 2000; McWhirter, 2000

Role of Schools in the Development of Aspirations

Role of schools in the Development of Aspirations

Wong, 1997; Paul, 1997; Orfield, 1997; Rosenbaum, 1997

Influence of coursework and curriculum tracking

Kelly & Lee, 2001; McWhirter, 2000; Mortimer, 2002; Kenny, 2003; Okagaki, 2001

Influence of coursework and curriculum tracking (cont'd)

Alexander & Entwisle, 2000; Johnson & Mortimer, 2000; Entwisle & Alexander 1993; Orfield, 1997; Paul, 1997; Kerckhoff, 1995; Lee & Bryk, 1988; Berends, 1992; Gamoran & Berends, 1987; Hallinan, 1996

Student perception of school supports/barriers

Daniels & Araposthasis, 2005; Klem & McConnell, 2004; Montalvo, 2007; Wehlage, 1989

●impact on engagement Yazzie-Mintz, 2009 ●classroom environment &

motivation Patrick, Kaplan & Ryan, 2007; Eccles & Wigfield, 1985; Ryan & Patrick, 2001

Role of Schools in the Development of Aspirations (cont'd)

●teacher support Goodenow, 1993; Skinner & Belmont, 1993; Ryan & Patrick, 2001; Schoon, 2001; Helwig 2004; Franklin, 1995; Flowers, Milner & Moore, 2003

●classroom organization, pedagogical methods

Ryan & Patrick, 2001; Barr & Dreeben, 1983; Gamoran & Berends, 1987; Gamoran & Mare, 1989; Hoffer, 1992; Kerckhoff, 1986, 1993; Murphy & Hallinger, 1989

School guidance counselors Hughes & Karp, 2004 ●history of; change in

role over time Hughes & Karp, Carey & Dimmitt, 2008

●importance of career development skills

Janosz, 1997; Turner, 2007

●perspective of /impact on students

Paul, 1997

●impact on student achievement and

occupational development

Borders & Drury, 1992; Fouad, 1995; Gerler, 1985; Lapan, 2004, 2007, 1993, 1997; Whiston &

191

Sexton, 1998

●impact on student aspirations and achievement

Rothney, 1958; McWhirter, 2000

●access to Mau, 1995; CEEB, 1986; Boyer, 1983; Lee & Ekstrom, 1987

Aspirationally ambitious teens - generational comparison

Schneider & Stevenson, 1999; Reynolds, 2006; Yazzie-Mintz, 2009; Goyette, 2008

The Problem of Misalignment Definition of misalignment Csikszentmihalyi & Schneider, 2000; Orfield, 1997; Rosenbaum, 2001; Schneider & Stevenson, 1999; Paul 1997

Indicators of Yazzie-Mintz, 2009; Csikszentmihalyi & Schneider, 2000; Rosenbaum, 2001; Stinchcombe, 1965; Kenny, 2003

Structural contribution to Rosenbaum, 2001; Reynolds, 2006; Johnson & Mortimer, 2002; Rosenbaum, Miller & Krie, 1996; Orfield, 1997; Paul, 1997; Stern, 2009; STWOA, 1994; Glover & Marshall, 1993; Mortimer, 2002; Modell, 1989; Gottfredson, 2005

The Problem of Misalignment (cont'd)

In relation to college Goyette, 2008; Resnick & Wirt, 1996; Rosenbaum, 1975, 1976, 1986, 2001; Schneider & Stevenson, 1999

Establish false foundation for future success

Rosenbaum, 1997, 2001; Schneider & Stevenson, 1999; NCES, 1999; Deil-Amen & Rosenbaum, 2002; Miller & Rosenbaum, 1997; Rosenbaum, 1996

Societal impact of Yazzie-Mintz, 2009; Lee & Rojewski, 2009; NCEE, 1983; Weiss, 2003; Sum, 2009; Rosenbaum, 2001; Haller & Portes, 1973; Blau & Duncan, 1967; Paul, 1997

Agency/self concept important for outcomes

Seifert, 2004; Mortimer, 2005; Csikszentmihalyi & Schneider, 2000; Super, 1953; Schoon, 2001; Kerckhoff, 1972

Contextual interactions impact self concept

Gottfredson, 1981; Kenny, 2003; Swanson & Woitke, 1997; Mortimer, 2005; Rotter, 1954

192

Social learning theory Rotter, 1954 Importance of Locus of Control

Definition of locus of control Rotter, 1966; Phares, 1976; Findley & Cooper, 1983; Mirowsky & Ross, 2003; Schieman & Plickert, 2008; deCharms, 1968; Graham, 1991; Seifert, 2004; Weiner, 1984

Importance to educational and occupational outcomes

Ryan & Patrick, 2001; Ames, 1992; Coleman, 1966; Coleman & DeLeire, 2003; Stipek, 1980; Ross & Broh, 2000; Murasko, 2007

The Importance of Locus of Control (cont'd)

External/Internal Rotter, 1966; Seifert, 1997, 2004; Rumberger & Palardy, 2004; Gifford, 2006; Rojewski, 1996; Mortimer, 2002; Glasgow, 1997; Murasko, 2007; Weiner, 1984; Csikszentmihalyi & Schneider, 2000; Hanson, 1994; Mau & Bikos, 2000; Bartels, 1971; Battle & Rotter, 1963; Shaw & Uh, 1971; Stipek, 1980

Schools as socializing context Csikszentmihalyi & Schneider, 2000; Eccles & Roeser, 2009; Eccles & Midgley, 1989

School Impact on Student Locus of Control

●tracking impact on locus Alexander & Entwisle, 2000; Gamoran, 2000; Cook, 1996; Kerckhoff, 1976, 1993, 1995, 2000; Rosenbaum, 1976; Gamoran, 1993; Gamoran & Berends, 1987; Slavin, 1990; Gamoran, 2000; Paul, 1997; Eccles & Midgley, 1989; Helwig, 2004

●access to info/counseling impact on

Johnson & Mortimer, 2002; Bradley & Gaa, 1977; Omizo & Omizo, 1988; Whiston & Sexton, 1998

●other school climate variables impact on (instruction,

press, mentoring)

Plucker 1998; Stipek, 1980; Johnson, 2000; Schultheiss, 2005; Super 1953, 1980; Turner, 2007; Deci & Ryan ,2000; Seifert & O'Keefe, 2001; Neild, 2009; Rumberger, 1995

193

REFERENCES

Alexander, K. L., & Entwisle, D. R. (2000). Educational tracking during the early years: First grade placements and middle school constraints. In A. C. Kerckhoff (Ed.), Generating social stratification: Toward a new research agenda (pp. 75-105). Boulder, CO: Westview Press.

Alexander, K. L., & McDill, E. L. (1976). Selection and allocation within schools: Some causes

and consequences of curriculum placement. American Sociological Review, 43, 963-980. Alliance on Excellent Education (AEE). (2009). The high cost of high school dropouts: What the

nation pays for inadequate schools Issue Brief. Washington D.C.: Alliance for Excellent Education.

Alwin, D. F. (1994). Aging, personality and social change: The stability of individual differences

over the adult life-span. In D. Featherman, R. Lerner, & M. Perlmutter (Eds.), Life-span development and behavior. (pp. 135-185). Hillsdale, NJ: Erlbaum.

Alwin, D. F., & Otto, L. B. (1977). High school context effects on aspirations. Sociology of

Education, 50(4), 259-273. Ames, C. (1992). Classrooms: Goals, structures and student motivation. Journal of Educational

Psychology, 84(3), 261-271. Armstrong, P. I., & Crombie, G. (2000). Compromises in adolescents' occupational aspirations

and expectations from grades 8 to 10. Journal of Vocational Behavior, 56, 82-98. Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.

Psychological Review, 84, 191-215. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory.

Englewood Cliffs, NJ: Prentice-Hall. Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social

psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173-1182.

Barr, R., & Dreeben, R. (1983). How schools work. Chicago, IL: The University of Chicago

Press. Bartel, N. (1971). Locus of control and achievement in middle and lower-class children. Child

Development, 42, 1099-1107.

194

Battle, E., & Rotter, J. B. (1963). Children's feelings of personal control as related to social class and ethnic group. Journal of Personality, 31, 482-490.

Becker, G. S. (1964). Human capital: A theoretical and empirical analysis with special reference

to education. New York, NY: Columbia University Press. Berends, M. (1992). Effects on school orientations and achievement in transition from middle to

high school. Ph.D. Dissertation. Department of Sociology, University of Wisconsin. Madison.

Berkner, L. (2000). Using National Educational Longitudinal Study data to examine the

transition to college. New Directions for Institutional Research, 107, 103-107. Blau, P. M., & Duncan, O. D. (1967). The American occupational structure. New York, NY:

Wiley. Borders, L. D., & Drury, S. M. (1992). Comprehensive school counseling programs: A review

for policymakers and practitioners. Journal of Counseling and Development, 70, 487-498.

Bourdieu, P. (1973). Cultural reproduction and social reproduction. In R. Brown (Ed.),

Knowledge, education and cultural change (pp. 71-112). London: Tavistock. Bourdieu, P., & Passeron, J. (1977). Reproduction in education, society and culture. London:

Sage. Boyer, E. L. (1983). High school: A report on secondary education in America. New York, NY:

Harper & Row. Bradley, R. K., & Gaa, J. P. (1977). Domain specific aspects of locus of control: Implications for

modifying locus of control orientation. Journal of School Psychology, 15(1), 18-24. Bransford, J. D. (1993). Who ya gonna call? Thoughts about teaching problem solving. In P.

Hallinger, K. Leithwood, & J. Murphy (Eds.), Cognitive perspectives on educational leadership. (pp. 171-191) New York, NY: Teachers College Press.

Bransford, J. D., Brown, A. L., & Cocking, R. R. (Eds.). (1999). How people learn: Brain, mind,

experience, and school. Washington, DC: National Academy Press. Bureau of Labor Statistics (2012). Number of jobs held, labor market activity, and earnings

growth among the youngest baby boomers: Results from a longitudinal survey. News Release. Washington, D.C.: U.S. Department of Labor.

Caravello, S. J. (1958). Effectiveness of high school guidance services. The Personnel and

Guidance Journal, 5, 323-325.

195

Carbonaro, W. (2005). Tracking, students' effort, and academic achievement. Sociology of Education, 78(1), 27-49.

Carey, J., & Dimmitt, C. (2008). A model for evidence-based elementary school counseling:

Using school data, research, and evaluation to enhance practice. The Elementary School Journal, 108(5), 422-430.

Cebi, M. (2007). Locus of control and human capital revisited. Journal of Human Resources,

42(4), 919-932. College Entrance Exam Board (CEEB). (1986). Commission on Precollege Guidance and

Counseling. Keeping the options open: Recommendations. New York, NY: College Entrance Exam Board.

Choy, S. (2002). Access & persistence: Findings from 10 years of longitudinal research on

students. Washington DC: Americal Council on Education. Committee on Institutional Cooperation (CIC). (1994). National Education Longitudinal Study

of 1988 CD-ROM Technical Documentation: Champaign, IL: Committee on Institutional Cooperation.

Cicourel, A. V., & Kitsuse, J. I. (1963). The educational decisionmakers. Indianapolis, IN:

Bobbs-Merrill. Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the

behavioral sciences. Hillsdale, NJ: Erlbaum Associates. Coleman, J. S., Campbell, E. Q., Hobson, C. J., McPartland, F., Mood, A. M., Weinfeld, F. D., &

York, R. L. (1966). Equality of educational opportunity. Washington, DC: US Department of Health, Education and Welfare, US Office of Education.

Coleman, M., & DeLeire, T. (2003). An economic model of locus of control and the human

capital investment decision. Journal of Human Resources, 38, 701. Cook, T. D., Church, M. B., Subira, A., Shadish, W. R., Kim, J., & Cohen, R. (1996). The

development of occupational aspirations and expectations among inner-city boys. Child Development, 67(6), 3368-3385.

Csikszentmihalyi, M., & Schneider, B. (2000). Becoming adult: How teenagers prepare for the

world of work (1st ed.). New York, NY: Basic Books. Curtin, T., Ingels, S., Wu, S., & Heuer, R. E. (2002). NELS 1988/2000: Base year to fourth

follow-up data user's manual. Washington, DC: US Department of Education, National Center for Educational Statistics.

196

Daniels, E., & Araposthasis, M. (2005). What do they really want?: Student voices and motivational research. Urban Education, 40(1), 34-59.

deCharms, R. (1968). From pawns to origins: Toward self-motivation. In G. Lesser (Ed.),

Psychology and educational practice. Glenview, IL: Scott Foresman. Deil-Amen, R., & Rosenbaum, J. (2002). Unintended consequences of stigma-free remediation.

Sociology of Education, 75(3), 249-268. Dennehy, K., & Mortimer, J. T. (1993). Work and family orientations of contemporary

adolescent boys and girls. In J. C. Hood (Ed.), Men, work and family (pp. 87-107). Newbury Park, CA: Sage.

Dewey, J. (1900). The school and society. Chicago, IL: University of Chicago Press. Dewey, J. (1897). My pedagogic creed. New York, NY: Teacher's College Press. Duncan, O. D. (1961). A socioeconomic index for all occupations. In A. J. Reiss Jr. (Ed.),

Occupations and social status (pp. 109-138). New York, NY: Free Press of Glencoe. Eccles, J. S., & Midgley, C. (1989). Stage/environment fit: Developmentally appropriate

classrooms for early adolescents. In R. Ames & C. Ames (Eds.), Research on motivation in education (Vol. 3, pp. 139-181). New York, NY: Academic Press.

Eccles, J. S., & Roeser, R. W. (2009). Schools, academic motivation and stage-environment fit.

In R. Lerner & L. Steinberg (Eds.), Handbook of adolescent psychology: Individual bases of adolescent development (3rd ed., Vol. 1). Hoboken, NJ: Wiley.

Eccles, J. S., & Wigfield, A. (1985). Teacher expectations and student motivation. In J. B. Dusek

(Ed.), Teacher expectancies (pp. 185-220). Hillsdale, NJ: Erlbaum. Entwisle, D. R., & Alexander, K. L. (1993). Entry into school: The beginning school transition

and educational stratification in the United States. Annual Review of Sociology, 19, 401-423.

Eyler, J., & Giles, D. E. (1999). Where's the learning in servicelLearning? (1st ed.). San

Francisco, CA: Jossey-Bass. Findley, M. J., & Cooper, H. M. (1983). Locus of control and academic achievement: A

literature review. Journal of Personality and Social Psychology, 44(2), 419-427. Flowers, L. A., Milner, H. R., & Moore, J. L. (2003). Effects of locus of control on African

American high school seniors' educational aspirations: Implications for preservice and inservice high school teachers and counselors. The High School Journal, 39-50.

197

Fouad, N. A. (1995). Career linking: An intervention to promote math and science career awareness. Journal of Counseling and Development, 73, 527-533.

Fouad, N. A. (2007). Work and vocational psychology: Theory, research and applications.

Annual Review of Psychology, 58, 543-564. Franklin, M. (1995). The effects of differential college environments on academic learning and

student perceptions of cognitive development. Research in Higher Education, 36, 127-153.

Freidlin, B., & Salvucci, S. (1995). Empirical evaluation of social, psychological and educational

construct variables used in NCES surveys (pp. 84). Washington, DC: National Center for Education Statistics.

Gamoran, A. (1993). Alternative uses of ability grouping in secondary schools: Can we bring

high-quality instruction to low-ability classes? American Journal of Education, 101, 1-22.

Gamoran, A. (1996). Student achievement in public magnet, public comprehensive, and private

city high schools. Education Evaluation and Policy Analysis, 18(1), 1-18. Gamoran, A. (2000). Educational stratification and individual careers. In A. C. Kerckhoff (Ed.),

Generating social stratification: Toward a new research agenda (pp. 59-74). Boulder, CO: Westview Press.

Gamoran, A., & Berends, M. (1987). The effects of stratification in secondary schools: Synthesis

of survey and ethnographic research. Review of Educational Research, 57, 415-435. Gamoran, A., & Mare, R. D. (1989). Secondary school stratification and educational inequality:

Compensation, reinforcement, or neutrality? American Journal of Sociology, 94, 1146-1183.

Gerler, E. R. (1985). Elementary school counseling research and the classroom learning

environment. Elementary School Guidance & Counseling, 20, 39-48. Gifford, D., Mianzo, F., & Briceno-Perriott, J. (2006). Locus of control: Academic achievement

and retention in a sample of university first year students. Journal of College Admission, 18-25.

Ginzberg, E., Ginsberg, S. W., Axelrad, S., & Herma, J. L. (1951). Occupational choice: An

approach to a general theory. New York, NY: Columbia University Press. Glasgow, K. L., Dornbusch, S. M., Troyer, L., Steinberg, L., & Ritter, P. (1997). Parenting

styles, adolescents' attributions, and educational outcomes in nine heterogeneous high schools. Child Development, 68(3), 507-529.

198

Glover, R. W., & Marshall, R. (1993). Improving the school-to-work transition of American adolescents. In R. Takanishi (Ed.), Adolescence in the 1990's: Risk and opportunity. New York, NY: Teachers College Press.

Goodenow, C. (1993). Classroom belonging among early adolescent students: Relationships to

motivation and achievement. Journal of Early Adolescence, 13, 21-43. Gottfredson, L. S. (1981). Circumscription and compromise: A developmental theory of

occupational aspirations. Journal of Counseling Psychology, 6(28), 545-579. Gottfredson, L. S. (2005). Applying Gottfredson's theory of circumspection and compromise in

career guidance and counseling. In S. D. Brown & R. W. Lent (Eds.), Career development and counseling: Putting theory and research to work (pp. 72-100). Hoboken, NJ: Wiley.

Goyette, K. A. (2008). College for some to college for all: Social background, occupational

expectations, and educational expectations over time. Social Science Research, 37, 461-484.

Goyette, K. A., & Mullen, A. (2006). Who studies the arts and sciences? Social background and

the choice and consequences of undergraduate field of study. The Journal of Higher Education, 77(3), 497-538.

Graham, S. (1991). A review of attributional theory in achievement contexts. Educational

Psychology Review, 3, 5-39. Haller, A. O., & Portes, A. (1973). Status attainment process. Sociology of Education, 46(1), 51-

91. Haller, A. O., & Woefel, J. (1972). Significant others and their expectations: Concepts and

instruments to measure interpersonal influence on status aspirations. Rural Sociology, 37(4), 591-622.

Hallinan, M. T. (1996). Track mobility in secondary school. Social Forces, 74, 983-1002. Hallinan, M. T. (2000). Educational processes and school reform. In A. C. Kerckhoff (Ed.),


Hansen, T. D., & McIntire, W. G. (1989). Family structure variables as predictors of educational

and vocational aspirations of high school seniors. Research in Rural Education, 6(2), 39-49.

Hanson, S. L. (1994). Lost talent: Unrealized educational aspirations and expectations among US

youths. Sociology of Education, 67(3), 159-183.

199

Hauser, R. M., & Feathermen, D. L. (1977). The process of stratification: Trends and analysis. New York: Academic Press.

Hauser, R. M., & Warren, J. R. (1997). Socioeconomic indexes for occupations: A review,

update, and critique. Sociological Methodology, 27, 177-298. Helwig, A. A. (2004). A ten-year longitudinal study of the career development of students:

Summary findings. Journal of Counseling and Development, 82(1), 49-57. Heyns, B. (1974). Social selection and stratification within schools. American Journal of

Sociology, 79, 1434-1451. Hodge, R. M., Seigel, P. M., & Rossi, P. H. (1964). Occupational prestige in the United States,

1925-1963. American Journal of Sociology, 70, 286-302. Hoffer, T. (1992). Middle school ability grouping and student achievement in sciences and

mathematics. Education Evaluation and Policy Analysis, 14, 205-228. Holland, J. L. (1985). Making vocational choices: A theory of vocational personalities and work

environments. Odessa, FL: Psychological Assessment Resources. Hughes, K. L., & Karp, M. M. (2004). School-based career development: A synthesis of the

literature: Institute on Education and the Economy. New York, NY: Teachers College, Columbia University.

Iacobucci, D. (2008). Mediation analysis. Thousand Oaks, CA: Sage. Ingels, S. J., Dowd, K. L., Baldridge, J. D., Stipe, J. L., Bartot, V. H., & Frankel, M. R. (1994).

User's manual: National Education Longitudinal Study of 1988 (NELS): Second follow up: Student component data file manual. Washington, DC: National Center for Educational Statistics.

Inoue, Y. (1999). The educational and occupational attainment process: The role of adolescent

status aspirations. Lanham: University Press of America. Janosz, M., LeBlanc, M., Boulerice, B., & Tremblay, R. E. (1997). Disentangling the weight of

school dropout predictors: A test on two longitudinal samples. Journal of Youth and Adolescence, 26, 733-762.

Jencks, C., Cocoran, M., Bartlett, S., Crouse, J., Englesfield, D., McClelland, K., . . . Williams, J.

(1979). Who gets ahead? New York, NY: Basic Books. Johnson, L. S. (2000). The relevance of school to career: A study in student awareness. Journal

of Career Development, 26, 163-171.

200

Johnson, M. K., & Mortimer, J. T. (2002). Career choice and development from a sociological perspective. In D. Brown (Ed.), Career choice and development (4th. ed., pp. 37-81). San Francisco, CA: Jossey-Bass.

Kalmijn, M. (1994). Mother's occupational status and children's schooling. American

Sociological Review, 59, 257-275. Kelly, K. R., & Lee, W. (2001). Mapping the domain of career decision problems. Journal of

Vocational Behavior, 61, 302-326. Kenny, M. E., Blustein, D. L., Chaves, A., Grossman, J. M., & Gallagher, L. A. (2003). The role

of perceived barriers and relational support in the educational and vocational lives of urban high school students. Journal of Counseling Psychology, 50(2), 142-155.

Kerckhoff, A. C. (1972). Socialization and social class. Englewood Cliffs, NJ: Prentice Hall. Kerckhoff, A. C. (1974). Ambition and attainment: A study of four samples of American boys.

Washington DC: American Sociological Association. Kerckhoff, A. C. (1976). The status attachement process: Socialization or allocation? Social

Forces, 55, 368-381. Kerckhoff, A. C. (1986). Effects of ability grouping in British secondary schools. American

Sociological Review, 51, 842-858. Kerckhoff, A. C. (1993). Diverging pathways: Social structure and career deflections.

Cambridge, MA: Cambridge University Press. Kerckhoff, A. C. (1995). Institutional arrangements and stratification processes in industrial

societies. Annual Review of Sociology, 21, 323-347. Kerckhoff, A. C. (2000a). Building conceptual and empirical bridges between studies of

educational and labor force careers. In A. C. Kerckhoff (Ed.), Generating social stratification: Toward a new research agenda (pp. 37-56). Boulder, CO: Westview Press.

Kerckhoff, A. C. (Ed.). (2000b). Generating social stratification: Toward a new research

agenda. Boulder, CO: Westview Press. Kerckhoff, A. C., & Huff, J. L. (1974). Parental influence on educational goals. Sociometry, 37,

307-327. Klem, A., & Connell, J. (2004). Relationships matter: Linking teacher support to student

engagement and achievement. Journal of School Health, 74(7), 262-273. Krull, J. L., & MacKinnon, D. P. (1999). Multilevel mediation modeling in group based

intervention studies. Evaluation Review, 23, 418-444.

201

Kulikowich, J. M., & Edwards, M. N. (2007). Analyzing change in school psychology research.

Psychology in the Schools, 44(5), 535-542. Lampard, R. (1995). Parents' occupations and their children's occupational attainment: A

contribution to the debate on the class assignment of families. Sociology, 29, 715-729. Lapan, R. T. (2004). Career development across the K-16 years: Bridging the present to

satisfying and successful futures. Alexandria, VA: American Counseling Association. Lapan, R. T., Aoyagi, M., & Kayson, M. (2007). Helping rural adolescents make successful

postsecondary transitions: A longitudinal study. ASCA Professional School Counseling, 10(3), 266-272.

Lapan, R. T., Gysbers, N., Hughey, K., & Arni, T. J. (1993). Evaluating a guidance and language

arts unit for high school juniors. Journal of Counseling and Development, 71, 444-451. Lapan, R. T., Gysbers, N., & Sun, Y. (1997). The impact of more fully implemented guidance

programs on the school experiences of high school students: A statewide evaluation study. Journal of Counseling and Development, 75, 292-302.

Lee, I. H., & Rojewski, J. W. (2009). Development of occupational aspiration prestige: A

piecewise latent growth model of selected influences. Journal of Vocational Bahavior, 75(1), 82-90.

Lee, J. M., & Rawls, A. (2010). The college completion agenda 2010 progress report: New York,

NY: The College Board Advocacy and Policy Center. Lee, R. B., & DeVore, I. (1975). Man the hunter. Chicago, IL: Aldine. Lee, S.M., Daniels, M.H., Puig, A., Newgent, R.A., & Nam, S.K. (2008). A data-based model to

prodeict postsecondary educational attainment of low-socioeconomic-status students. ASCA Professional School Counseling 11, 306-316.

Lee, V. E., & Bryk, A. S. (1988). Curriculum tracking as mediating the social distribution of

high school achievement. Sociology of Education, 61, 78-94. Lee, V. E., & Bryk, A. S. (1989). A multilevel model of the social distribution of high school

achievement. Sociology of Education, 62(3), 172-192. Lee, V. E., Bryk, A. S., & Smith, J. B. (1993). The effects of high school organization on

teachers and students. In L. Darling-Hammond (Ed.), Review of research in education (pp. 171-268). Washington DC: American Educational Research Association.

Lee, V. E., & Ekstrom, R. B. (1987). Student access to guidance counseling in high school.

American Educational Research Journal, 24, 287-310.

202

Lent, R. W., Brown, S. D., & Hackett, G. (1996). Career development from a social cognitive

perspective. In D. Brown, L. Brooks & Associates (Eds.), Career choice and development (3rd ed., pp. 423-475). San Francisco, CA: Jossey-Bass.

Lent, R. W., Brown, S. D., & Larkin, K. C. (1986). Self-efficacy in the prediction of academic

performance and perceived career options. Journal of Counseling Psychology, 33, 265-269.

Little, R., & Rubin, D. (1987). Statistical analysis with missing data. New York, NY: Wiley. Maas, C. J. M., & Hox, J. (2004). The influence of violations of assumptions on multilevel

parameter estimates and their standard errors. Computational Statistics & Data Analysis, 46, 427-440.

MacKinnon, D. P. (2008). Introduction to statistical mediation analysis. New York, NY:

Erlbaum. MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002). A

comparison of methods to test mediation and other intervening variable effects. Psychological Methods, 7, 83-104.

Macleod, J. (1987). Ain't no making it. Boulder, CO: Westview Press. Marcia, J. E. (1987). The identity status approach to the study of ego identity development. In T.

Honess & K. Yardley (Eds.), Self and Identity (pp. 161-171). New York, NY: Routledge & Kegan Paul.

Marjoribanks, K. (1985). Families, schools, and aspirations: Ethnic group differences. Journal of

Experimental Education, 53, 141-147. Mau, W.-C. (1995). Educational planning and academic achievement of middle school students:

A racial and cultural comparison. Journal of Counseling and Development, 73(5), 518-526.

Mau, W.-C., & Bikos, L. H. (2000). Educational and vocational aspirations of minority and

female students: A longitudinal study. Journal of Counseling and Development, 78, 186-194.

McClelland, K. (1990). Cumulative disadvantage among the highly ambitious. Sociology of

Education, 63, 102-121. McWhirter, E. H., Rasheed, S., & Crothers, M. (2000). The effects of high school career

education on social-cognitive variables. Journal of Counseling Psychology, 47(3), 330-341.

203

Merton, R. K. (1957). Social theory and social structure. New York, NY: Free Press. Meyer, J. (1977). The effects of education as an institution. American Journal of Sociology, 83,

55-77. Miller, L., & Bud, J. (1999). The development of occupational sex-role sterotypes, occupational

preferences and academic subject prefrences in children at ages 8, 12 and 16. Educational Psychologist, 19, 17-35.

Miller, S. R., & Rosenbaum, J. (1997). Hiring in a hobbesian world: Social infrastructure and

employers' use of information. Work and Occupations, 24(4), 498-523. Mirowsky, J., & Ross, C. E. (2003). Social causes of psychological distress (2nd ed.). Hawthorne,

NY: Aldine De Gruyter. Modell, J. (1989). Into one's own: From youth to adulthood in the United States 1920-1975:

Berkeley:University of California. Montalvo, G. P., Mansfield, E. A., & Miller, R. B. (2007). Liking or disliking a teacher: Student

motivation, engagement and achievement. Evaluation and Research in Education, 20(3), 144-158.

Mortimer, J. T. (2000). Social psychological aspects of achievement. In A. C. Kerckhoff (Ed.),


Mortimer, J. T., Staff, J., & Lee, J. C. (2005). Agency and structure in educational attainment

and the transition to adulthood. Advances in Life Course Research, 10, 131-153. Mortimer, J. T., Vuolo, M., Staff, J., Wakefield, S., & Xie, W. (2008). Tracing the timing of

"career" acquisition in a contemporary youth cohort. Work and Occupations, 35(1), 44-84.

Mortimer, J. T., Zimmer-Gembeck, M. J., Holmes, M., & Shanahan, M. J. (2002). The process of

occupational decision making: Patterns during the transition to adulthood. Journal of Vocational Bahavior, 61, 439-465.

Murasko, J. E. (2007). A lifecourse study on education and health: The relationship between

childhood psychosocial resources and outcomes in adolescence and young adulthood. Social Science Research, 36, 1348-1370.

Murnane, R. J., Willett, J. B., & Levy, F. (1995). The growing importance of cognitive skills in

wage determination. Review of Economics and Statistics, 77(2), 251-266. Murphy, J., & Hallinger, P. (1989). Equity as access to learning: Curricular and instructional

treatment differences. Journal of Curriculum Studies, 21, 129-149.

204

Nakao, K., & Treas, J. (1992). The 1989 socioeconomic index of occupations: construction from

the 1989 prestige scores: General Social Survey. N.p. Nakao, K., & Treas, J. (1994). Updating occupational prestige and socioeconomic scores: How

the new measures measure up. Sociological Methodology, 24, 1-72. National Commission on Excellence in Education (NCEE). (1983). A nation at risk: The

imperative for educational reform: A report to the nation and the secretary of education, United States Department of Education. Washington, DC: National Commission on Excellence in Education, Superintendent of Documents, USGPO.

National Center for Education Statistics (NCES). (1996a). National Educational Longitudinal

Study of 1988 (NELS). Restricted file. Washington, DC: National Center for Education Statistics, US Department of Education.

National Center for Education Statistics (NCES). (1996b). National Educational Longitudinal

Study of 1988-1994, Data Files and Electronic Codebook System. Washington, DC: Office of Educational Research and Improvement, US Department of Education.

National Center for Education Statistics (NCES). (1999). Digest of Educational Statistics.

Washington, DC: National Center for Education Statistics, U.S. Department of Education.

National Center for Education Statistics (NCES). (2000). Electronic Codebooks for Windows

95/98. Washington, DC: Office of Educational Research and Improvement, US Department of Education.

National Center for Education Statistics (NCES). (2006). Economic outcomes of high school

completers and noncompleters 8 years later. Washington, DC: National Center for Education Statistics, US Department of Education.

National Center for Education Statistics (NCES). (2009). High school dropout and completion

rates in the United States: 2007. Washington, DC: National Center for Education Statistics, US Department of Education.

Neild, R. C. (2009). Falling off track during the transition to high school: What we know and

what can be done. The Future of Children, 19(1), 53-78. No Child Left Behind Act of 2001 (NCLB), Pub. L. No. 107-110, 1425 Stat. (2002). O'Rand, A. (2000). Structuration and indivdualization: The life course as a continuous,

multilevel process. In A. C. Kerckhoff (Ed.), Generating social stratification: Toward a new research agenda (pp. 3-16). Boulder, CO: Westview Press.

205

Oakes, J., Gamoran, A., & Page, R. N. (1992). Curriculum differentiation: Opportunities, outcomes and meanings. In P. W. Jackson (Ed.), Handbook of research on curriculum (pp. 570-608). Washington DC: American Educational Research Association.

Okagaki, L. (2001). Triarchic model of minority children’s school achievement. Educational

Psychologist, 36, 9-20. Omizo, M. M., & Omizo, S. A. (1988). Group counseling's effect on self-concept and social

behavior among children with learning disabilities. Journal of Humanistic Education and Development, 26, 109-117.

Orfield, G. (1997). Going to work: Weak preparation, little help. In K. Wong (Ed.), Advances in

Educational Policy, 3, 3-31. Greenwich, CT: JAI Press. Paisley, P. O., & Hayes, R. L. (2003). School counseling in the academic domain:

Transformations in preparation and practice. Professional School Counseling, 6, 198-204. Parcel, T. L., & Dufur, M. J. (2001). Capital at home and at school: Effects on student

achievement. Social Forces, 79(3), 881-911. Patrick, H., Ryan, A. M., & Kaplan, A. (2007). Early adolescents' perceptions of the classroom

social environment, motivational beliefs, and engagement. Journal of Educational Psychology, 99(1), 83-98.

Paul, F. G. (1997a). Aspirations and outcomes: The mediating role of information, preparation,

and paths. In K. Wong (Ed.), Advances in educational policy (Vol. 3, pp. 93-136). Greenwich, CT: JAI Press.

Paul, F. G. (1997b). Negotiated identities and academic program choice. In K. Wong (Ed.),

Advances in educational policy (Vol. 3, pp. 43-78). Greenwich, CT: JAI Press. Phares, E. (1976). Locus of control in personality.Morristown, NJ: General Learning Press. Pintrich, P. R. (2000). Educational psychology at the millennium: A look back and a look

forward. Educational Psychologist, 35(4), 221-226. Pintrich, P. R., & De Groot, E. V. (1990). Motivational and self-regulated learning components

of classroom and academic performance. Journal of Educational Psychology, 82(1), 33-40.

Pituch, K. A., Murphy, D. L., & Tate, R. L. (2010). Three-level models for indirect effects in

school- and class-randomized experiments in education. The Journal of Experimental Education, 78, 60-95.

Plucker, J. A. (1998). The relationship between school climate conditions and student

aspirations. The Journal of Educational Research, 91(4), 240-246.

206

Rasinski, K. A., Ingels, S. J., Rock, D., Pollack, J. M., & Wu, S. (1993). America's high school

sophomores: A ten year comparison. Washington, DC: US Department of Education. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data

analysis methods (2nd ed.). Thousand Oaks, CA: Sage. Raudenbush, S. W., Bryk, A. S., Cheong, Y. F., Congdon, R., & du Toit, M. (2004). HLM 6:

Hierarchical linear and nonlinear modeling. Lincolnwood: SSI: Scientific Software International.

Reiss, A. J. (1961). Occupations and social status. New York, NY: Free Press of Glencoe. Resnick, L. B., & Wirt, J. G. (1996). The changing workplace. In L. B. Resnick & J. G. Wirt

(Eds.), Linking school and work: Roles for standards and assessment (pp. 12-31). San Francisco, CA: Jossey-Bass.

Reynolds, J., Stewart, M., Macdonald, R., & Sischo, L. (2006). Have adolescents become too

ambitious? High school seniors' educational and occupational plans, 1976-2000. Social Problems, 53(2), 186-206.

Rindfuss, B. S., Cooksey, E. C., & Sutterlin, R. L. (1999). Young adult occupational

achievement: Early expectations versus behavioral reality. Work and Occupations, 26, 220-263.

Robertshaw, D., & Wolfle, L. M. (1980). Discontinuities in schooling and educational

attainment. Paper presented at the American Educational Research Association, Boston, MA.

Rock, D. A. (2012). Modeling change in large-scale longitudinal studies of educational growth:

Four decades of contributions to the assessment of educational growth (ETS Research Report No. RR 12-04). Scientific and Policy Contributions Series No. SPC 12-01. Princeton, NY: Educational Testing Service.

Rojewski, J. W. (1996). Educational and occupational aspirations of high school seniors with

learning disabilities. Exceptional Children, 62, 463-476. Rojewski, J. W. (2005). Occupational aspirations: Constructs, meanings and application. In S. D.

Brown & R. W. Lent (Eds.), Career development and counseling: Putting theory and research to work (pp. 131-154). Hoboken, NJ: Wiley.

Rojewski, J. W., & Yang, B. (1997). Longitudinal analysis of select influences on adolescents'

occupational aspirations. Journal of Vocational Behavior, 51, 375-410. Rosenbaum, J. (1975). The stratification of socialization processes. American Sociological

Review, 40(1), 48-54.

207

Rosenbaum, J. (1976). Making inequality: The hidden curriculum of high school tracking. New

York, NY: Wiley. Rosenbaum, J. (1986). Institutional career structures and the social construction of ability. In J.

Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 139-172). Westport: Greenwood Press.

Rosenbaum, J. (1998). College for all: Do students understand what college demands? Social

Psychology of Education, 2(1), 55-80. Rosenbaum, J. (2001). Beyond college for all: Career paths for the forgotten half. New York,

NY: American Sociological Association. Rosenbaum, J., Milller, S. R., & Krie, M. S. (1996). Gatekeeping in an era of more open gates:

High school counselors views of their influence on students' college plans. American Journal of Education, 104(4), 257-279.

Rosenbaum, J., Miller, S. R., & Krei, M. A. (1997). What role should counselors have? A

commentary on "Negotiated identities and academic program choice". In K. K. Wong (Ed.), Advances in Educational Policy (Vol. 3, pp. 79-91). Greenwich, CT: JAI Press.

Rosenbaum, J., Miller, S. R., & Roy, K. (1996). Long-term effects of high school grades and job

placements. Paper presented at the American Sociological Association, New York, NY. Ross, C. E., & Broh, B. A. (2000). The roles of self-esteem and the sense of personal control in

the academic achievement process. Sociology of Education, 73(4), 270-284. Rossi, P. H., Lipsey, M. W., & Freeman, H. E. (2004). Evaluation: A systematic approach (7th

ed.). Thousand Oaks, CA: Sage. Rothney, J. W. M. (1958). Guidance practices and results. New York, NY: Harper. Rotter, J. B. (1954). Social learning and clinical psychology. Englewood Cliffs, NJ: Prentice-

Hall. Rotter, J. B. (1966). Generalized expectancies for internal versus external control of

reinforcement. Psychological Monographs, 80(1, Whole No. 609), 1-28. Rubin, D. (1976). Inference and missing data. Biometrika, 63, 581-592. Rubin, D. (1987). Multiple imputation for nonresponse in surveys. New York, NY: Wiley. Rumberger, R. W. (1995). Dropping out of middle school: A multilevel analysis of students and

schools. American Educational Research Journal, 32(3), 583-625.

208

Rumberger, R. W., & Palardy, G. J. (2004). Multilevel models for school effectiveness research. In D. Kaplan (Ed.), Handbook of qualitative methodology for the social sciences: Thousands Oaks, CA: Sage.

Rury, J. L. (2002). Democracy’s high school? Social change and American secondary education

in the post-Conant era. American Education Research Journal, 39(2), 307-336. Ryan, A. M., & Patrick, H. (2001). The classroom social environment and changes in

adolescents' motivation and engagement during middle school. American Educational Research Journal, 38(2), 437-460.

Ryan, R., & Deci, E. (2000). When rewards compete with nature: The undermining of intrinsic

motivation and self-regulation. In C. Sansone & J. Harackiewicz (Eds.), Intrinsic and extrinsic motivation: The search for optimal performance. New York, NY: Academic Press.

Savickas, M. L., Nota, L., Rossier, J., Dauwalder, J.-P., Duarter, M. E., Guichard, J., . . . van

Vianen, A. (2009). Life designing: A paradigm for career construction in the 21st century. Journal of Vocational Bahavior, 75(3), 239-250.

Schieman, S., & Plickert, G. (2008). How knowledge is power: education and the sense of

control. Social Forces, 87(1), 153-183. Schneider, B., & Stevenson, D. (1999). The ambitious generation: America's teenagers

motivated but directionless. New Haven, CT: Yale University Press. School-to-Work Opportunities Act (STWOA), Public Law 103-239, 108 Stat 568, May 4, 1994,

(20 U.S.C.6 101-6235) (expires October 1, 2001). Schoon, I. (2001). Teenage job aspirations and career attainment in adulthood: A 17 year follow-

up study of teenagers who aspired to become scientists, health professionals or engineers. International Journal of Behavioral Development, 25(2), 124-132.

Schoon, I., & Parsons, S. (2002). Teenage aspirations for future careers and occupational

outcomes. Journal of Vocational Behavior, 60, 262-288. Schulenberg, J., Vondracek, F. W., & Crouter, A. C. (1984). The influence of the family on

vocational development. Journal of Marriage and Family, 10, 129-143. Schultheiss, D. E. P. (2005). Elementary career intervention programs: Social action initiatives.

Journal of Career Development, 31(3), 185-194. Seifert, T. L. (1997). Academic goals and emotions: Results of a structure equation model and

cluster analysis. British Journal of Educational Psychology, 71, 323-338. Seifert, T. L. (2004). Understanding student motivation. Educational Research, 46(2), 137-149.

209

Seifert, T. L., & O'Keefe, B. (2001). The relationship of work avoidance and learning goals to

perceived competency, externality and meaning. British Journal of Educational Psychology, 71, 81-92.

Sewell, W. H., Haller, A. O., & Ohlendorf, G. W. (1970). The educational and early

occupational status attainment process: Replication and revision. American Sociological Review, 35(6), 1014-1027.

Sewell, W. H., Haller, A. O., & Portes, A. (1969). The educational and early occupational

attainment process. American Sociological Review, 34(1), 82-92. Shanahan, M. J. (2000). Pathways to adulthood in changing societies: Variability and

mechanisms in life course perspective. Annual Review of Sociology, 26, 667-692. Shaw, R., & Uhl, N. (1971). Control of reinforcement and academic achievement. The Journal of

Educational Research, 64, 226-228. Skinner, E. A., & Belmont, M. J. (1993). Motivation in the classroom: Reciprocal effects of

teacher behavior and student engagement across the school year. Journal of Educational Psychology, 85, 571-581.

Slavin, R. E. (1990). Achievement effects of ability grouping in secondary schools: A best-

evidence synthesis. Review of Educational Research, 60, 471-499. Smith-Maddox, R. (1999). The social networks and resources of African American eighth

graders: Evidence from the national educational longitudinal study of 1988. Adolescence, 34, 169-183.

Sorokin, P. A. (1927). Social mobility. New York, NY: Harper. Steinberg, L., Dornbusch, S. M., & Brown, B. B. (1992). Ethnic differences in adolescent

achievement: An ecological perspective. American Psychologist, 6, 723-729. Stern, D. (2009). Expanding policy options for educating teenagers. The Future of Children,

19(1), 211-239. Stevens, G., & Cho, J. H. (1985). Socioeconomic indexes and the new 1980 census occupational

classification scheme. Social Science Research, 14, 142-168. Stevens, G., & Hoisington, E. (1987). Occupational prestige and the 1980 U.S. labor force.

Social Science Research, 16, 74-105. Stinchcombe, A. L. (1965). Social structure and organizations. In J. G. March (Ed.), Handbook

of organizations.(pp. 142-193). New York: Rand McNally.

210

Stipek, D. (1980). A causal analysis of the relationship between locus of control and academic achievement in first grade. Contemporary Educational Psychology, 5, 90-99.

Strong, E. K. (1953). Validity of occupational choice. Educational and Psychological

Measurement, 13, 110-121. Sum, A., Khatiwada, I., McLaughlin, J., & Palma, S. (2009). The consequences of dropping out

of high school: Joblessness and jailing for high school dropouts and the high cost for taxpayers. Boston, MA: Center for Labor Market Studies, Northeastern University.

Super, D. E. (1953). A theory of vocational development. American Psychologist, 8, 185-190. Super, D. E. (1980). A life-span, life-space appraoch to career development. Journal of

Vocational Behavior, 16, 282-298. Swanson, J. L., & Woitke, M. B. (1997). Theory into practice in career assessment for women:

Assessment and interventions regarding perceived barriers. Journal of Career Assessment, 5, 443-462.

Trice, A. D., & McClellan, N. C. (1993). Do children's career aspirations predict adult

occupations? An answer from a secondary analysis of a longitudinal study. Psychological Reports, 72, 368-370.

Trusty, J., Niles, S. G., & Carney, J. V. (2005). Education-career planning and middle school

counselors. ASCA Professional School Counseling, 9(2), 136-142. Turner, S. L. (2007). Preparing inner-city adolescents to transition into high school. ASCA

Professional School Counseling, 10(3), 245-252. Updegraff, L. A. (1996). Gender roles in marriage: What do they mean for girls' and boys' school

achievement. Journal of Youth and Adolescence, 35, 78-88. Wang, L., Kick, E., Fraser, J., & Burns, T. J. (1999). Status attainment in America: The roles of

locus of control and self-esteem in educational and occupational outcomes. Sociological Spectrum, 19, 281-298.

Wehlage, G. G., Rutter, R. A., Smith, G. A., Lesko, N., & Fernandez, R. R. (1989). Reducing the

risk: Schools as communities of support. London: Falmer. Weiner, B. (1984). Principles for a theory of student motivation and their application with an

attributional framework. Research on Motivation in Education: Student Motivation, 6, 15-38.

Weiss, S. (2003, December 15). Highlights from the 2003 National Forum on Education Policy:

Nation at risk continues to affect education system: Education Commission of the States.

211

Whiston, S. C., & Sexton, T. L. (1998). A review of school counseling outcome research: Implications for practice. Journal of Counseling and Development, 76, 412-426.

William T. Grant Foundation, (1988). The forgotten half: Pathways to success for America’s

youth and young families. Final report. Washington DC: Commission on Work Family and Citizenship.

Woefel, J. (1972). Significant others and their role relationships to students in high school

populations. Rural Sociology, 37, 86-97. Wong, K. (1997a). Advances in educational policy (Vol. 3). Greenwich, CT: JAI Press. Wong, K. (1997b). Preface: Meeting the challenge of workforce preparation. In K. Wong (Ed.),

Advances in educational policy (Vol. 3, pp. ix-xvii). Greenwich, CT: JAI Press. Yazzie-Mintz, E. (2009). High school survey of student engagement. Bloomington: Indiana

University School of Education, Center for Evaluation & Education Policy Zhang, Z., Zyphur, M. J., & Preacher, K. J. (2009). Testing multilevel mediation using

hierarchical linear models. Organizational Research Methods, 12(4), 695-719.